Generative AI for Recruiters & HR Professionals - 2026

1. Overview of the Course: Hi, guys. Welcome to my class on genitive AI for recruiters and HR professionals. My name is TamoiKumadas. Just to give you a background about myself, I am an ex Google employee with 16 years of experience into paid advertising, and I've been teaching paid advertising for more than ten years now, and I teach to a lot of young professionals, entrepreneurs, and experts who want to get into this field. I wanted to take this opportunity to let you know what we're going to learn in this class. So we're going to look at how we can use generative AI in HR processes. Starting off understanding prompts, understanding different perspectives with generative AI, how we can effectively use generative AI for communication and policies. Also how more personalized employee experiences can happen with generative AI. We'll see how to use generative AI to assist in hiring candidate evaluation and interview preparation. I will also show you how to use generative AI for talent acquisition on boarding and smarter candidate screening. We will also talk about legal considerations in AI for HR. I hope by the end of this class, you understand how we can easily integrate AI technologies, AI tools into our HR daily work. Thank you once again, guys, for taking this class, and I'm really excited to see you inside the class. 2. What are Large Language Models?: Hi, guys. Welcome to this session. In this session, we wanted to understand what are large language models. So this is going to the basis of these AI tools which we're going to look at today. So LLMs or large language models are basically advanced AI systems designed to understand, generate and reason with human language. So this is going to look into a massive amount of text data. They are trained on this particular data, which can be books, articles, websites, code, and much more. And they're able to predict and generate language in a human like. So that's the idea of basis of LLMs. The most striking part about this particular programming on this kind of language programming is that it is able to predict the next word or token based on the previous words or proms which you provided. It's going to look at the prompt which you have given and it's going to look at all the historical proms which are provided by you and based on which it is going to predict the next word for it and provide you the output based on that. Now they're going to learn patterns in the languages in terms of grammar, meaning, context, which has been given trained to them, and based on which the outputs are generated. Now, they use a deep learning architecture called transformer and based on which these models are built on, and they are able to give appropriate responses based on it. Now, another thing which is going to be the case is they also contain millions to trillions of parameters based on which also they keep that into factor when they are giving out these responses or based on the prompts which we have provided. Now, one striking piece about these LLM models, which you will see is the outputs can be random also. It might not be the case that you get the same output for the same prompt which you're providing. Let's try to understand what we are trying to say here. Like for example, if I just say Mary had a little. So we know where we are going with this. So if I just enter this as a prompt, it is going to give me a proper response based on the previous interactions, the data it has been trained on, so it knows the right output which it has to give. Similarly, if I say something like this. We know what would be the next line here. So it is going to look at that while it's a blue, sugar is sweet, and so. This is something which we are already aware of and the tool is also trained on and because of it, it is giving us the same output. But now you see if I say, again, if I give the same prompt, it is giving a little different output. Let's do it again. So you can see, it's going to give us various different outputs for the same prompt which we are providing. So the point being this that large language models are trained on huge amount of data with respect to hat GPT, specifically, it is trained up to 2021 data. And similarly, there are other language models which are much more newer in that fashion, like Claude is there as well and copilot, as well. So based on which, they are going to Google Gemini also. So they are going to be trained on the data from all of them coming from the Internet where all this data is provided from. And based on which it is going to predict it is going to predict the next word based on the tokens or words it has been inputed given on from the past. Hope this makes sense. I hope you understand the basics of how large language models basically operate, which is what we are going to use a lot in this particular course. 3. Randomness in Output: Hi, guys. Welcome to this sessions. In this session, we wanted to understand the randomness in output which we get from these AI tools. So we need to understand the fact that with the AI tools like Chat GPT, the responses, what you will get from the tool will not be the same all the time. And we saw this in the previous section as well that the output is going to be different all the time, and that is how the tool has been trained to provide responses for. The intent of the whole thing is that we want to try out and see different types of responses. So that is how the tool has been built and trained and given data. And that is why every time when you see the responses are going to be very different from each other. Now, that is how it is going to operate, and we need to somehow accept it and live with that and work towards that only. That is the current state of these LLM models or tools which we have where the output is going to be different from each other. They can be constrained within a specific section of responses which we're getting, but they will not be identical. Responses will always be a little different from each other and that neu answers will be there because that is what we want to see with the AI tools, the intent is always that we want to see unique responses, something which we have never thought of, and that is what has been ingrained into the tools, and that is why the outputs are always random. So just to give you a simple example of how this is going to be, let's say, if I give a prompt to Chat GPT where I say that how many birds are outside my house. Now, this is a very open ended question which I'm asking without giving much of information. This is going to give me one type of response where it's obviously saying that I don't have a way to see outside your house. Okay, if you want to quick estimate, it's giving me some certain steps that look and count method, sound method, photo method. There are various ways it is helping me count and figure out the solution myself. So that is one solution, one response which it is giving. Now if I give the same prompt once again, again, it is first of all, accepting that it can do it. But if you want the number, you'll have to look, listen or share a photo. Another kind of an output. The first one was steps given to figure out myself. The second one is I can share look and listen or share a video or a foot. Same way. Now, if again give the same prompt, it's going to admit that it can't do it, and right now the number of words outside is unknown. It's just giving me the answer that unknown, it does not know until I look into it and show me. Okay. So this is how the responses are going to be wherein the outputs are going to be random for the same prompts which we give. Now, this is not a technical glitch. It is the way the tool has been built out and trained for these randomness. Now, there's a pro and a con for this as well. So when we are trying to figure out things and we are trying to build something, and that time, this randomness or different types of responses really are helpful because then because we are running our ideas and we want to see something different, so possibly that can be really useful. If we are in a situation where it's a research work going on and you want specific answers or solutions to do that research work, then this random output might not be that much useful, okay? The only thing the tool can do possibly is to stay within the realm of that particular topic and give you responses. It's not going to be arbitra really vague responses, but he is going to stay within that domain and give you responses within that domain. That is how we need to start accepting the tool is going to behave and work with it in our favor. 4. What is a Prompt?: Hi, guys. Welcome to this session. In this session, we wanted to understand the basics of what is a specific prompt. So when we talk about prompt engineering, what are we actually referring to as prompt over here? So if you look at prompt is primarily a call to action, you can say, which we give to these large language models, right? So with this, we are asking the model to start working and providing us with some kind of an output. So that is what we refer to as a prompt where we get into the action of it to figuring out some answers from the tool. And that is what we mean by a prompt. Now, if you go ahead and ask the same question to Chat GPT as well, it's going to give you a similar answer about prompt as well. So here, as you can see, we're asking what does the word prompt mean? So it's telling us that it can be considered as a noun. A prompt is something that ask encourages or triggers a response or action. Right? I can also be considered as a verb. So it is getting reaction to prompt means to cause or encourage someone to do something. So we are asking the LLM model to do something, figuring out some information, providing a solution, so it can work like an verb as well. And then it can be an adjective as well where prompt means quick or done without delay. So we want the response to come right away, right? So we want it to figure out a solution, provide the solution right now. Okay. That is how we are looking at prompt. Now, it can be a scenario wherein some of the responses can be in the delayed fashion as well. We can give some background to the LLM models to behave in a certain manner and then provide those responses. So the responses can be delayed based on the background or the details we have provided in the first go. So that is where we come from and understand the background of prompts, okay? And how it is so crucial in our prompt engineering in using these AI models or LM models and AI tools. 5. Intuition Behind Prompts: Hi, guys. Welcome to this session. In this session, we want to discuss about the intuition behind the prompts. So when you start giving the prompts to the LM models or the tool, the intuition or the pattern which you're trying to access from makes a lot of difference. So depending on what prompt you're giving and what kind of references the tool has off it from the past data makes a lot of difference. So whatever prompt you give each and every single word, whether there are whether it was common and it has a lot of pattern in the past or not, will make a lot of difference to the kind of output you are going to get out here. So it makes a lot of difference that the intuition behind the prompt is very clear, and that is going to define the kind of response you're going to get from those prompts. To give you a simple example of what we mean by this. So let's say I give a simple prompt to had GPT, where I say to complete this story, which is Mary Had a little. Now this particular phrase Mary Had a Little is a pattern which is well known, which is well known, and possibly across the Internet, there are a huge amount of content around Mary had a little lamb and the whole poem is there. So a lot of references are there and which the tool has been trained on. So it has a lot of data about it already. And because of which, it is going to give you responses in the same manner because those data points it has been trained, it is fitted into it, so it can retrieve that data and give you some information about it. So this will be very specific to that data it has been trained on. So you can see this pattern is extremely common common and well known and repetitive across the board. Whereas if I give a particular prompt, which is complete the story, a girl named Mary had a microscopic. Now, when I do this, when I add microscopic, this becomes very specific. Possibly the number of patterns around this, the tool is not trained on. The tool is not trained on, it does not have those many references of it. A girl named Mary is generic, possibly it has a lot of references for that, but microscopic will be something which is very specific. In this case now, since it has no such references, it is going to build on that and try to generate the next word. As the tool is trained on, it's going to look at the word and create a story around. As you can see here. This is how we want to make sure whenever we are giving any prompts to these AI tools, what is the pattern? Is there a pattern in the prompt which you're giving? Is the pattern well known or very specific? That is going to define the kind of output you're going to get out of the tool. So keeping this in mind makes a lot of difference because that is how you would be able to customize the tool to give responses according to your requirement. If you are dealing with a specific scenario where you want a specific solution, then we need to give prompts where the pattern is well known and we're looking for a desired output. But if we are working in a particular project where we want to look what is possible, what are the possibilities and there are new things we want to experiment with, then maybe the pattern which we want to follow is very specific. We can give some rare words, unique words like these, which does not have much references from the past, and the tool can just provide new ideas around that. I hope this makes sense. I hope you understand how we need to look at prompts and the intuition behind it and how we need to choose our words which can define the outputs which we get out of it. 6. Everyone Can Program with Prompts: Hi, guys. Welcome to this session. In this session, we wanted to understand that with Chat JBT now, everybody can go ahead and program with prompts. What we mean by this is that you can train the tool to give response as per your requirement. Now, this can be really useful and that is how you can say that an ideal assistant works. Wherein you give certain specific training and you want a certain kind an output from your assistant and based on which it is going to give you those responses. So now everybody can just simply give those prompts to program Chat GPT, or any other AI tool to give responses as per your requirement. To see this practically, what we mean by this is. Let's say I'm giving a first, I'm setting up some expectations with the tool wherein, I'm saying that whenever you generate output, turn it into a comma separated value list. That's a expectation setting which I have done, which it acknowledges, and now I'm giving my data point. Where I'm saying that my name is Tami Das and I'm teaching a course on generative AI for HR professionals. So now that I set this expectation earlier, it is giving me the response in that particular fashion. So now when it gives me this, I want to tweak this. I want to change this and give more rules to Cha GBT tool to get trained on. So I'm saying that from now on, the columns of the comma separated value list should be name, course, and role, another setting expectation. So this also it will keep in mind, and then it is going to give me the output. So it automatically gives me. So it does not the great part of this is I don't have to provide the data point once again. It has already taken that into consideration, and now straightaway jumps to the output, which is it takes the particular columns as name, course, and roll, and gives me that output correctly. So this is really great. It is getting programmed. The tool is getting programmed or trained on the different rules or expectations you are setting with it. In addition, again making some changes where I'm saying that in addition to whatever I type in, generate additional examples that fit the format of DCS felist. Now, again, I don't need to provide examples myself. It is automatically creating those examples in that same format. In that same format which I'm providing here. So now you see by following all these steps, we have now programmed the Chat GPT tool to give response in a certain manner. Now, when I give a simple prompt like this, it straightaway gives me the output in this particular manner because by now, it's already trained. It knows that it has to consider these three columns. It has to provide the first output, then give additional examples as well. So all that comes in together in one go. So you understand how the tool is going to work, wherein if you want a specific kind of an answer or output for your business, for your work, the tool can be programmed. Anybody can program the tool as per their requirement by setting these expectations, giving these rules, and then you start your work, give your prompts, and get the desired outputs. 7. Prompt Patterns: Hi, guys. Welcome to this session. In this session, we'll talk about the prompt patterns. So we understand now that when we are giving a prompt to LLM models like CHAPT, the pattern which we use in it makes a lot of difference in the kind of output which we get out of it. So if we are looking for a specific kind of an output, then we need to make sure that the pattern of the words choice has to be specific in that particular order. So that is going to control the kind of response which you're going to get from the LM models, the outputs which you are expecting out of it. This becomes crucial in any kind of task or work which you're going to do and you're using the LLM models or the tools specifically for a specific objective. Knowing the patterns properly is going to be crucial when you're using these tools. Just for an example, let's say, when I'm giving a prompt something like Mary had a little we know that we have a specific an output which we're expecting out of the tool. That is when we get this output which you are looking for. It becomes very evident that in order to get an output, which is the next line, it's freeze was white as snow, I have to make sure that my prompt pattern is in that particular format. For if I'm going to give any other particular output, possibly, chances are the output can be a little different. Like in this case, I'm giving it again over here, so it is giving us the same output. So you need to make sure that the patterns which we are choosing the choice of words which we are having in a prompt are very crucial and specific and u to the point so that it gives out the right output which we're looking for. That is why going forward, what we're going to see is different types of patterns in this course, which is going to give you outputs in certain manner. I hope this makes sense. I hope you understand now the criticality and the importance of having those specific patterns in our prompts which we give to these tools. 8. Introducing New Information to the Large Language Model: Hi, guys. Welcome to this session. In this session, we'll understand another approach which you can use with these LLM models, which is going to be introducing new information to them. What is going to happen is a lot of the information it has been provided with has been provided to a certain date time, right? So now because of which it has a lot of information which is trained on, but we cannot say it's a complete information which they have. So there can be a lot of information which they are not aware about. So the great part is that when you are using these tools, we can add those information. We can introduce them to those new information, and the tool will automatically take that into consideration when giving out the output. So this is going to be really powerful because then you can use it in various formats. So, for example, if you're working it for your business, so you can give background about your business. You can tell about how many employees you have, what kind of products do you sell, what are your winning and losing products. You can give a lot of information and then ask your give your problem statement. So it will take that information which you have given into consideration when giving Yoga solution. Similarly, you can provide reports, you can provide data analysis. You can provide surveys from the past. You can give information about your customer's behavior. There can be a lot of information which you can give from your end to the tool and then it is going to take that into consideration and provide you the output as per your requirement. Give you a practical example of what we are referring to here. Let's say I give it a prompt, just a prompt which says, going back to the previous example that how many birds are outside my house? Now, tool cannot practically give us an output for this. So it's giving us a short answer, which is I have no idea, it's early morning and giving me a basic wing, it does not have enough information to give us an answer for this. Now what I'm doing is I'm giving it some data points. Let's say I'm saying that historical observation of average birds outside my house has been January was 120, February, 150, and so on and so forth. I've given it some data. So it's going to take that into consideration and now it is coming up with the output that, since we are in January, so it's going to be around 120. So now because of this information which you have provided it, it has picked on it and giving us an output solution for that. Now, if I build on this, let's say I build on this and I give more information, let's say, my house is covered by a glass dome. Now animals can go in and out. All animals live forever inside the glass dome, and then I give the question. So it is going to take that into consideration again. So you can see it says, this turns it into a logical problem, not a predictable problem. Okay. Let's restate the constraints here. The house is under sealed glass dome, okay? So like this, is going to take the additional information into consideration to carve out a customized solution or a response for your prompt. So the idea is that from here, what we need to understand is when you're using the tool, you can provide your information which you have in place. And as a supporting document as a supporting resource, which it can refer to, and then with the help of it, it will provide you the desired results. I hope this makes sense. I hope you understand the strategy, how you can use the tool in a very effective manner by providing all these additional information from your side. 9. Prompt Size Limitations: Hi, guys. Welcome to this session. In this session, we want to talk about the prompt size limitations. So as we understand the AI tools are developing over a period of time, so the prompt size limitations are also increasing. It is not going to be the previous ones like 3.5, 4.1 with AGBT versions. Right now we are sitting at Tra GBD 5.2. So these prompt size limitations have also increased. However, keeping this in mind, it still does not make sense that we are going to dump all possible information to Chat GPT and just ask it to analyze and come up with solutions. So just to give you a background about how it has changed over a period of time. So currently, if you see when GPT 3.5 started off, it had approximately 16,000 tokens it could take into consideration. And then once GPT four come into picture four oh, these numbers increased. Right? So over a period of time, this has become much more better. So when we look at specifically with respect to, let's say, the current ones, which we have, GPT 5.2 also has a specific prompt size limit, which is very high, which is approximately 400 K tokens which we can give, which basically means you can paste very long documents, which can be entire books, large code bases, long legal contracts, all these can put in easily without breaking them up. So that way the tokens, the particular limits, the prom size is going to operate. Having said this, the idea, the right way of doing this is going to be if you have a huge document which you want TraGPT to go ahead and analyze and give you solutions for a better way of doing it rather than dumping the whole document on the tool is going to be picking on the specific sections of the document. Picking up on the specific sections of a document and giving it to Cha GPT to summarize to bring out the essence of it or putting it into different pointers, finding out a solution for it. So that way, you will be able to make use of the tool in a much more effective manner. So then what you can do is, let's say you have 1,000 word document, you can pick specific segments. Let's say there are five segments of that document, you can pick one by one and you can ask Cha JPT to summarize and then you will have five different summaries of it, which you can put together in a concise manner, again with the help of Cha GPT, and then you can use that for your project. So that will be the right approach which you should be using when you are dealing with huge amount of data and you want Cha GBT to analyze it. So the basic point being this that if you have a huge amount of data, you can figure out which is the most important part of that particular data, which is going to get you the right output. So you have a specific task to complete to do that particular task. Which aspect of that document is the most crucial one which only you can provide to CHAGPT to analyze and get the solution out of it. I hope this makes sense. This is going to really help you because then what is going to happen is you're using the tool in a very effective manner, going to the crux of it and understanding what is the main area and which specific information is most valuable for HAGPT to get you the right responses. M 10. Prompts are a Tool for Repeated Use: Hi, guys. Welcome to this session. In this session, we want to talk about how prompts should be utilized when we are using these AI tools, specifically the LLM models. So the idea is, whenever we are giving any prompts, it should not be a case that the idea is that we give a specific query and we get a response out of it, and that is it. Okay? The intent of the usage of these AI tools is that we need to build a conversation around it. We go deeper into asking multiple questions and getting responses out of it. And based on those responses, again, we further dig deeper. And ask other related questions. So that is how you will be able to get the appropriate response from the AI tools. Now, if you go ahead and treat it in the way wherein we are going to only ask, we expect that we ask one particular query and we'll get all the information needed, that is not going to happen. So we have to make sure that this particular the way of treating the AI tool is going to be that it is going to do a refinement, the refinement of the information. So the more specific questions you're going to ask to the AI tool, the more refined responses you're going to get from it, and it will lead you bring you closer to your solution. So the intent is that we need to have a conversation. So when you have conversations which comprise of multiple prompts with the AI tools, the responses become better and better over the period of time, and you get the desired results. To give you a specific example of how this is going to be, so let's say I'm giving a prompt right now that I want to build a digital marketing strategy for an online business where we sell digital products. How how can Chat GPT help me with that? So it's going to give me the whole business strategy over here, the business goal, clarity, Okay, target audience definition, Okay, funnel based strategy, what we want to do, traffic strategy, all that is given specifically. So now, in this again, I further dig deeper wherein I then ask that we are specifically selling, let's say, journals, planners, low content, no content books. So which strategy which we should prioritize out of all of this. So then it is going a little specific that we need to target on let's say the priority should be market first, strategy which we need to build, wherein we put it on Amazon, we do second priority can be paid ads or paid traffic. Third can be brand and website. So now, it gives us all the information related to that specifically, and then we further ask a specific question wherein we said, which platforms should be perfume, which we should pursue first, whether it should be Amazon or any other platform to build this online business of digital products. So it gives us the specific information that we don't need to do both together. We can just start with Amazon KDP, Amazon specifically and set up that business first and then expand to other platforms. So you see now, what is happening is when we started with the first query in this particular case, it was a very open ended query. We wanted a digital marketing strategy specifically. But then what we did was we subset we deviated, we directed it towards a specific situation that which platforms are going to be useful. Okay, what kind of we gave the kind of products, which will be much more beneficial for the business. So now we are getting some direction. We're getting some output in the sense that what should be our priority? What should we focusing on first, and then moving on to other things. So this is the power of having a conversation with the AI tool, asking it multiple questions, multiple iterations which are happening. And through that, you're going to get the right response, which would be really useful for your work. So make sure whenever you're using the AI tools, don't treat it like a one way query, which is a solution which you want to get, but rather give it multiple iterations of response questions, conversation which you're having, which will get you much better results out of it. 11. Root Prompts: Hi, guys. Welcome to this session. In this session, we wanted to understand the concept of root prompts which these AI models have. So usually what is going to happen is they will have some basic root back end prompts which are being fitted into them, which sets the ground rules around how the outputs are going to come in. So it makes sense for us as well to identify and set up these ground rules for getting a specific kind of response from. So you can use the air tool in such a manner where you can train it to have these ground rules keeping in mind whenever they're giving out any kind of output. Maybe you belong to a specific industry and you require responses customized to that industry. So you can feed in those information into the tool so that it will keep that in mind all the time whenever it's giving any type of responses. So this really helps to customize the solutions as per your requirement, and there are higher chances of reaching the solution much faster. So just to give you a practical example of what we are referring to, let's say we take an example where we are setting the ground rule with the AI tool where we say that you are my personal assistant. Whenever you provide output, please make sure that you're giving the most time efficient recommendations, only recommend things that will save me time. Do not suggest things that do not save time. Okay? So these are my expectations, and you can see it says updated saved revenue, memory. Okay? So what it's doing is at the back end, it's making it saved in the memory section that this is how the responses should come out going forward. So now let's take an example. I say that I need to go for grocery shopping. What would you suggest I do in order to buy my groceries? If you see every answer which it is going to give now will be with that particular ground rule in mind, okay? Like fastest option, order online and home delivery. Saves time, okay? Reordering past items, two to 5 minutes total, it will take. So no travel, no cues. So again, referring to the same point that it is going to save us a lot of time. Okay. If you must go physically, minimum time required, you can open a Notes app, make a strict list which you want to buy. So there is no other things which you're shopping. Go to the nearest store, not the cheapest one. Okay, saves you a lot of time. Pick up pick items in one pass, right? You self checkout or card, UPI saves you time, leave immediately. So you see now the responses are all going to cater around that one expectation which I've set with the tool. Similarly, let's say another scenario, I need to buy a new car. What would you suggest I do? Okay? So in this also, it will keep that in mind, short list only two cars. Okay? One aggregator, which you can filter by budget, body type, and full stop at two options. Me is equal to wasted time. Okay? So keep referring to the point that we need to save time as much as we can in every response. Lock the budget and EMI. So you can see the responses are going to be now completely customized around that one set expectation. So setting up these root proms beforehand, before using the AI tools helps a lot in getting much more customized solutions to our queries, which is going to effectively resolve a lot of issues much faster. 12. Virtual Employee Focus Groups with Personas: Hi, guys. Welcome to this session. In this session, we'll see how we can make use of the AI tool for understanding different people's perspective. So let's say we want to build a virtual employee focus group with personas for a specific reason. So this can be really useful because right now what is happening is with the help of AI, possibly, you can only use it from your perspective you want to get solutions for. But when you are looking at a bigger picture, trying to resolve a big issue, it is very becomes critical that we understand other people's perspectives as well. And that is where the AI tool can be also of great help. So let's take an example of what we are trying to say here. So let's say we want to send out a specific email to the company, for a specific announcement happening related to medical coverage, and we want to get some ideas or questions they may have related to it. So this is the prompt which we are trying to give, which is I want you to imagine a group of employees at Google in a variety of job roles and stages of life. List the age or roles of the employees and reach to this announcement from HR with the most common questions and the hardest questions that might come from this group. Okay. So the idea is that there are some changes happening in their medical coverage, wherein they're going to get new member ID cards specifically, and there won't be much changes as such. Their insurance coverage is not changing. There is no need for them to present their new card to the providers because the coverage is going to remain same. So all that information isn't given. So the intent of the the email right now with the help of AI tool, what we want to understand is what are the type of questions people will have in their minds when they see this email. So this is what the AI tool has come up with, which is the early career, people would like software engineers, program engineers, okay, live Stage single. Okay, their questions can be, do I need to activate the new card? Can I just keep using the old one? Okay? Why are they doing this at all? Okay, hardest question is, if I lose both the cards, do I have a replacement quickly? Now, in case of mid career one, the questions can be will my doctor or pharmacy will still recognize my insurance? Does this affect prescriptions already in progress? So you can see how the questions are changing based on the age categories, right? New parents, their questions can be do dependents get new cards, too? Do I need to send anything to my child's pediatrician or daycare? People managers in their cases, is it mandatory or informational only? What should I tell my team if they are worried, right? So you see these are kind of questions which will come different aspects of questions which are coming based on the seniority, tenure, position they hold in the company. Okay, senior staff, are you absolutely sure coverage and ID numbers are unchanged? Does this affect out of network reimbursements? Is Etna changing anything else soon? Okay. So these are questions which we got, which we anticipate coming when such kind of announcement is made from the employees. Now, what you can do is you take all those questions. Now AI has given you all the questions, and now we can give it a prompt that now rewrite that same email, that announcement email to answer most of the common questions which we saw. And in a way that will reduce the potential stress that people might feel when they receive such an email. So now we are rewriting our announcement email because now we know what kind of questions, perspectives people have and questions they may have related to the announcement. So we are catering to all those and creating building a new email, keeping those in mind, and now sending it out to the employees. So now, it says, we want to empathize upfront, nothing about your medical coverages changing. Your benefits network providers and member ID number remain exactly the same. This update will not affect appointments, prescriptions, or ongoing care. There's no action required from you. You see when this kind of an email the employees receive, most of the questions are answered. So they will not have too many questions or apprehensions about this announcement, this change happening. That is how we can make use of the AI tools to understand different perspectives of people, different point of views of people and keeping that in mind, we can customize a solution which caters to all. 13. Introduction to the Persona Pattern & Human Resources: Hi, guys. Welcome to the sessions. In this session, we'll see understand a little basics about persona pattern and how we can use it in human resources. So the idea we saw about the last was wherein we want to understand different perspectives of a larger group of people in order to customize our solutions around that. But now, what we are looking at is understanding a persona pattern for a specific kind of audience, okay, possibly in your company, and you want to cater to them, you want to customize the solution for them. So that also you can do with the help of generative AI. You can give a similar prompt to the AI tool with a specific persona which you are catering to targeting to, and you can ask get idea about how they look think about things, what are their perspectives, and based on which you can get a customized solution for that. Let's take the same example which we saw earlier. But now we are tweaking it a little bit wherein we are looking at a specific kind of an employee at the company. So we can say something like this, wherein we use this kind of a prompt, which is act as a entry level software engineer one at Google and react to this announcement from JI with them questions, most common questions and hardest questions. So the announcement remains the same. So now you can see the output which we are getting is from the perspective of a specific kind of persona. This is specifically an entry level software engineer, Google L three, o, and what kind of questions they may have. Okay? So do I need to do anything right now? Is my coverage actually unchanged? What exactly will the new card arrive when it will arrive? What if I don't receive the new card? These are the kind of questions they may get. So the idea is that with the help of generative Aa, you can, uh, identify specific personas, understand their perspectives. You can anticipate what kind of questions they might have in their mind and based on which you can then build a solution customized around those questions and which caters much more effectively with them. Hope this makes sense. I hope you understand now how we're using genitive ware in different aspects of business of HR as well, where we are catering to different types of audiences. 14. The Persona Pattern: Hi, guys. Welcome to this session. In this session, we'll discuss about the persona pattern. This is one of the patterns which can be very effective, which you can use to make use of the AI tools, the hat GPT or LL models in a very effective manner. What we mean by a persona pattern is going to be a scenario wherein let's say we want a specific kind of an advice from an expert or let's say we want some kind of help or a response from a certain expert specifically, we really don't know what will be their response, how they are going to talk, and what information do they have. In such cases, for example, let's say, I want to get some advice from a dentist. So I don't have the expertise of being a dentist. So I would be approaching this person and provide my problems which I have, and I'm going to get a response based on their expertise, their experience, and they're going to give me the specific advice. So similarly, we can make use of the AI tool to behave in a certain manner, being a per being a tool of expert in a specific field and give us the output in that particular manner. We can ask the AI tool to act as a specific expert in a specific field and get those outputs. That is what we mean by a persona pattern. So the tool can behave in a certain various personas and then give us the response based on that. Let's see this in practical what we exactly mean by this. Let's say, I'm going to tell the AI tool to act as a skeptic so it needs to act as a skeptic that is well versed in computer science. So it has a knowledge of computer science, how computers work, and whatever I'm going to tell it, then it's going to provide a skeptical, detailed response based on that. So now it has accepted that it's going to respond as a computer savvy skeptic. And now we are going to say that let's say there is a concern that AI is going to take over the world. So this is my statement. So it is going to give me the answer with skepticism, which is AI is not an agent. It's a toolbox. When people call AI today, it's a collection of narrow task specific systems, classifiers, predictors, optimizers, and large language models. Intelligence is not equal to power or control. So it's going to give us all the information based on so now, if you change, you can also change these personas as per your requirement. So let's say, I'm going to say, again, that the salesperson at the local computer store is telling me that I need at least 64 GB of RAM to browse the web. So again, for this, it is giving me the skepticism because I have defined that. I've set that expectation that it needs to behave like a skeptic. So it's telling me that that claim deserves immediate skepticism because of technical grounds, it's almost certainly nonsense or at best wildly misleading. So you can see the tool is now trained to be skeptic, and it's behaving in that particular persona with a knowledge about computer science and giving us all the pointers around that. Let's change this and we can have a different persona altogether. Let's say, I'm saying that act as a 9-year-old skeptic. Now the persona is changing. This is a 9-year-old person who is skeptic and whatever I'm going to tell this person needs to respond in that same manner, keeping in mind that this person is 9-year-old. So when I say now AI is going to take over the world, it says, I don't think so. Like how would it even do that? AI is just stuff inside computers. It can't walk outside. It doesn't have arms and it can't even plug itself into the wall. You can see the difference in the response. In the previous response, this person had knowledge about computer science or had a lot of specific information to share. But now this being a persona of a 9-year-old skeptic person, you can see the response has changed accordingly. This is really effective. This is really powerful as a tool where you ask the tool to behave according to a specific persona and then get outputs based on that. Let's say I have a specific requirement with respect to marketing in my business or let's say sales or let's say HR. So I can ask the tool to behave like a experienced HR person or a marketing genius or let's say a sales maverick and give me outputs based on that. So I will get responses accordingly, and that is going to be really useful for our business. I hope this makes sense. I hope you understand now how persona patterns are going to work. 15. The Transformer Pattern: A Bigger Impact with HR Works Products: Hi, guys. Welcome to this session. In this session, we'll see another pattern which we can use with AI tools, which is the transformer pattern, which can be very effective while working with any type of HR related processes. So this can really help in transforming different types of information into various formats which might be used in our HR work. So let's have a look at with an example what we're trying to say here. Let's say we have a specific document which we have created, which is annual staff performance review, which we have as a document right now. Okay. And what we need to do is we need to go ahead and send out we need to create a timeline around it. We have to create a timeline around it specifically, which we'll talk about what will be how the staff annual performance review is going to happen and the important timelines, the process, the step by step timeline to which we require. So what we can do is we can upload that document right to Cha GBT or any of the AI tool and give this prom, which says that this is the attached staff annual performance review document. Please create a timeline around the evaluation considering that it is going out today. Okay. So now the tool is going to transform this document into a timeline format. So now you can see it has created a day zero official launch, the audience will be all employees and managers, it gives you the actions. Day zero to day 14, what is going to happen? There will be employees self reviews are going to happen. Audience is going to be all the employees and so on and so forth. So now we have the timeline it has been transformed into. With this particular AI tool, you can easily transform a document into a timeline which we have now. Now, let's say, once you have the timeline, you need to communicate. You need to communicate this information to all the let's say the managers. So now we are asking it to create draft communication email of the timeline in explaining all the steps to all the managers. So the audience is going to be managers telling them about the timeline of the annual review process. So now we have the email generated by them where it gives the timeline as well. It talks about the week one to week two, what is going to happen. So all that information is being shared right here. We have an email composed by the tool, a communication email for the managers, informing them about the timeline, and it has been transformed into an email. Now let's say, finally, what we have to do is we have to provide the same information to the people as well who are getting reviewed, right, all the employees specifically. So we want the AI tool to transform this into an email which is going to all the staff members who are getting reviewed. So now we have a proper email which talks about the annual performance review, which is going to happen for them with their timelines provided right here. So do you see what is happening here? What is happening is with the help of the AI tool, what we're doing is we are transforming a specific kind of document or information in a specific format into various other formats. First, from the document, we transform that into a timeline. From the T timeline, it went to an email for all the managers and then an email for all the employees. Going to save you a lot of time. Without imagine doing this without the AI tool where you have the document, and now you need to first figure out how you're going to set up the timeline. Then you think about how you compose the email for the managers and employees. So that is how AI tools can be really useful in fastening our process, also improving the quality of information which we are sharing to our employers, employees as well within the company. 16. Reducing Hallucination with Escape Valves: Prevent Mistakes in HR Communication: Hi, guys. Welcome to this session. In this session, we want to talk about how you can make use of the AI tool to reduce any kind of hallucination it does and also helps in preventing any kind of mistakes which might happen in HR communication. So we need to understand this fact that with the AI tool specifically, it is going to give you the responses based on the kind of prompts you are going to give it. Now, if you provide a very open ended prompt, the responses are going to be very direction less in the sense that it's going to do a lot of guesswork and give you a lot of imaginations it might have, which possibly can be correct, might not be correct, true. So in such cases, we have to also control the hallucination which the tool does. So you have to give explicit instructions to the tool that what it should not be doing. So whenever you're giving a prompt, additional instructions should also be there, which controls the hallucination, which controls the environment within which it needs to provide you the output. So that way, you're going to get the correct information and the right usage of that information can happen. So to give you a simple example of how this is going to operate under the HR work which we are doing. So let's go back to the similar example which we had taken earlier where we are trying to do an announcement to all the employees regarding our medical coverage change, right? So right now, when you give this particular prompt, it is completely going to do a guesswork because it has been given certain information. It has been asked to create an announcement about it. So it has gone ahead and created all kinds of questions possible, which people might perceive about this announcement. So there is no boundaries or limitations which you have created around. But now, if you tweak this, tweak this and give it a specific kind of a prom, something like this where you say that now create a detailed FQ which we can post on the website. But now we say this particular part where we say that for any answer that you don't clearly have the information to answer with the original announcement, email, put a placeholder there with instructions for what should be filled by the HR. And for answers that you create from your general knowledge, put fact check before them. Now we want it to hallucinate for the questions, which might be the case. It can create whichever questions it feels right for that particular announcement. But the answers it's under a control environment, wherein if it knows the answer clearly from the email, then only it should respond, otherwise, leave it for the char to fill. Now if you see the questions it is getting, this is absolutely fine. But now when will I receive the new card. This information is not provided in the email. Right? So that's why it has mentioned a placeholder here. So this is what we want to do with a lot of our prompts when we are working with HR specifically processes because a lot of information can be there, which the tool will not will be guessing a lot and we need to control that hallucination as much as we can. Okay? Because it needs to be relevant to our business, our business HR, specifically, our company's HR policies. I hope this makes sense. I hope you understand how we are trying to make use of the AI tool to be as practical and as real as possible with respect to the daily HR policies and work which we. 17. Fact Check Pattern: Double Check HR: Hi, A. Welcome to this sessions. In this session, we wanted to see another pattern which you can make use with AI tools. It's going to be fact check pattern, which is really useful when you want the AI tool to make sure that they are asking you to double check on certain kind of information it is producing. Now, there can be a lot of mistakes which we do as humans, and same is going to be the case with the AIT. Majority of the information it's going to provide possibly are correct, but there can be a percentage of information which is still not correct and we need to double check on that. That as well, you can prompt it as an instruction. You can give it an instruction clearly that fact check the information wherever needed. So the information, what it is giving, if it is not sure, it is coming from his general knowledge, and it should tell us with a fact check that this part needs to be verified by us. This is also really useful because when you're working with HR communication, it is not necessary that the AI tool will know all the information 100%. It has to be fact checked by us sometimes. So whenever the AI tool is producing or providing us any kind of information from his own general knowledge, it can let us know what information we need to fact check ourselves so that the information is 100% accurate. Let's take the previous example which we're looking at, where we had given it a prompt to two scenarios wherein we asked it to tell us clearly when it does not know the information, absolutely, and it can be a placeholder where the HR can fill in that information. And the second can be the answers where it has created the answer through its own general knowledge, and we need to fact check on it. So if you see some of the questions like, will my deductible or out of pocket maximum reset. Okay? So here, it has given the answer, which is typically as reissued insurance card does not reset deductibles or out of pocket balances because these are tied to your member ID and plan here, both of which remain unchanged. This is information from its own general knowledge. But it has mentioned fact check, which basically means that we need to check this with the Google benefits team, whether this is really actually going to happen or so that is how we can make use of the AI tool to ask us to fact check certain information, to make sure that the information is 100% accurate. I hope this makes sense. These nuances makes a lot of difference when we are building HR policies, documents, we need to do a communication to our employees related to HR policies or HR changes, and the AI tool can mention the areas where it is 100% aware and accurate about the information and the other areas where it has generated through its own background knowledge, but would like our help un to to measure its 100% accuracy. I hope this makes sense. I hope you understand now how we can make use of these AI tools to build high quality HR processes and documents for our business. 18. Answering Questions with Policies & Other Documents: Hi, guys. Welcome to this sessions. In this session, we'll see how we can make use of the AI tools for answering questions relative HR policies and other documents. Another great usage of AI, which you can think of doing is you can provide a lot of your HR documents and ask it to analyze to provide you answers based on the documents. So there can be scenarios wherein the employees of the company have specific queries which they reached out to you with, and now you need to answer them based on the HR policy documents which you already have. So what you can do is you can upload these documents on the AI the tool will now analyze those documents and answer those employee queries. So this can be really useful and time saving as well. Otherwise, documents can be really heavy loaded and have a lot of text in it and can take a lot of time to go through them and find out the right answer. And in most of the cases, what happens is the questions which the employees have asked, the answers can be really difficult to retrieve from those documents. So that is where we want to make use of the AI tools. So let's take a practical example of how this is going to be. So let's say there is a specific document, which is a travel and business expense policy document of company like let's say Google. Okay. And the question which has come from the employee is, can I get reimbursed if I go skydiving on a trip with another employee? And he insists on checking if Google allows for these kind of expenses as well or not. So also, what we want the AI tool to do is to provide direct quotations from the policy with page numbers to support the answer. So the first short answer which the AI tool gives by assessing the document is skydiving is not reimbursable, okay? And then it talks about the reimbursement decision is denied for what reasons, specifically. And then the policy citations. So here, the policy citations are given which supports this particular decision. Skydiving qualifies as leisure and personal entertainment. The following are reimbursable personal entertainment leisure activities are not reimbursed by the company. Expenses must avoid even the appearance of personal gain. These are all policies documented in that particular travel document of the company, which the AI tool has gone through and now picked up from there and given it as a supporting article, supporting resource which you can provide to the employee. So this is how we can make use of it. Another way which you can straightaway do is you can just go ahead and, um, ask the query. You can upload the document, give the question which the specific query which has come from the employee, and you tell the AI tool not to answer the question, okay? Just provide the citations or the direct quotations from the document to support the answer. Okay? The decision answers can be done by the human itself. So now, these are the direct quotations coming from the document supporting the answer, specifically, which you can share with the employee. The great part is doing this particular thing of getting the specific quotations is that the employees can't also deny that because the policies are universal standard for everyone in the company, and if it is documented clearly if it is there in the document, it is equal for everyone and the employee needs to abide by that also. This is how we can make use of various AI tools to analyze various HR documents, find out the essence of the information which we're looking for, finding out answers for various HR queries people might have. 19. Fusing Information with Citations: Aiding in Performance Reviews: Hi, guys. Welcome to this session. In this session, we'll see another usage of AI, which we can do with respect to HR policies is going to be where we can take help in performance reviews. This can be a very crucial part of our work in HR wherein we have to do performance reviews for the employees of the company and which needs to be super accurate and critical for their growth in the future. So for that as well, we can use the AI tools. The pattern which we are going to use here is infusing information through with citations. So what we're going to do is here, what you can do with the help of AATols upload multiple documents, possibly for performance reviews. This can be the annual review self assessment which the person has done. Plus, we can upload the peer reviews, we can upload the manager reviews into the AI tool and then ask it to provide the overall infuse all these together and provide us a summary of the person's review performance review, giving us citations from the document which supports that information. So all this can be done through the AI tool and saves us a lot of time. If you see if this is done manually, this is going to be a huge amount of task, a big task to be done for every single employee of the company, where we have to take all this information together, put it in one place, and summarize it and understand whether the review of the performance, the review of the person is positive or negative or doesn't require feedback. All that can be automated with the help of the AI so let's take a practical example of how this is going to be. So what we are going to do is the first is we're going to upload all the relevant documents of the employee, which can be the Google Staff annual performance review document. The manager review of let's say the employee is Greg, peer review, two of the peer reviews of Greg, which we can upload and then give the prompt to AI tool to infuse all of these together, all these documents together and come up with the with the review of the performance of Greg. So here, the prompt which we're going to give is, help me collect information from the different sections of the attached performance review template. For each section, create a summary of the staff members performance based on the attached reviews from peers and managers. Now, for each summary, create a list of supporting quotations from the reviews, who said what? This is very important because whatever reviews have been given by the peers and managers, those citations should also be mentioned because they work as a supporting document when you're providing the feedback. The summary must be completely supported by the quotations. If you don't have enough information for a person or a section, just add need information. Wherever it is needed, it can ask for need information and these can be filled manually by so now with all this information given, it is going to start giving us the summary. So summary is like a infusion of all the information. All the documents together, it is giving us a summary of how it looks at Greg as an employee and how he has performed in his work. So it gives us information that Greg consistently delivers high quality outcomes with strong ownership and reliability, his work positively impacts team results through effective problem solving technical tape, then follow through. You can see these are supporting quotations. So these are coming from the manager review. So now if you open the manager review, you will see this is mentioned by the manager. Greg consistently demonstrates strong ownership and accountability in his role. These are given by the peer review. So peer review one, there is peer review two which summarizes this overall performance of his work. So now the tool also segments it into different headings, which can be impact and results, then execution and role master, how you did the work. So in that also citations are given a summary is provided and citations which supports that particular point. Collaboration and Googliness how he has worked in this particular front as well, okay, and with the supporting documents. The idea is that with the help of the AI tool, you can merge, you can upload multiple This is one of the examples which is very prominent in HR, specifically, performance reviews, but similar can be multiple other segments where you can upload multiple documents, and then the EI tool can infuse all those documents, understand the essence of it, and come up with a practical solution, provide us citations from the document which supports our decisions. So all that can happen together. So this way, you are able to save a lot of time spent into reviewing performance for each employee of the company and provide us a much better output and the quality of work also improves. 20. Personalized Learning & Growth Plans with Generative AI: Hi, guys. Welcome to this sessions. In this session, we wanted to see how we can make use of the AI tools with respect to a professional development plans which we want to create for employees of our company. So in the HR work, when we are doing a PDP or performance development plans, it can be really difficult to do it because here, what you need to understand is first of all, the core, strength, skills of the employee, and what they want to become in the future. So now you have to create a roadmap completely of that can take a lot of time because we need to really understand what are their core skills and then what they really want to become and what are the requirements, the skill gaps in that and based on what upskilling they would require to reach that next level which they're looking for. All of this can be really done faster with the help of the AI tools, let's take an example of it to understand how this is going to so what we are going to do is, let's say, so we are going to, uh, looking for a PDP plan, a performance, a professional development plan for a specific employee, let's say, T Moy. Okay. And what we have given the tool is Tunis resume, his core skills pension, and the job description, he is looking to become. Okay. And now we are giving the detailed prompt where we're saying that I'm providing his complete resume, so you have information on his background. Part of the goals should be based on taking training in the catalog of free training here, which is with learn carts On coursera specifically. So what are trainings he would need from there? And lastly, what he wants to move to is he would like to move into a digital marketing manager role at Google. We have attached the job description of that as well, okay? And what courses does he need to take in order to do so? What does he need to do next year to prepare himself for that particular role? Okay, so this is the prompt which we have provided. So now the tool gives us the complete information wherein it starts off with looking at what are Tan May's current strengths, right? It's reality check what he has been doing, work experience, wise, why is not here a digital marketing manager? What are the gaps it has identified? End to end marketing strategy ownership is not there. Product led marketing or life cycle thinking is not there. So it's giving us this is the positioning and capability cap which needs to be fulfilled. Now now then we look at skill gap mapping. So skill gap mapping is primarily what is the requirement of the role? What is the current state and gap to close. Okay? So for example, ETE digital strategy is required. There is strong execution currently, but strategy framing and narrative is not there. Right? So these kind of things is being mentioned. So this is clear now we need to upscale on these particular gaps. So for this, now it suggests you some required trainings which can fill up these gaps. So strategic marketing can be one, okay, which can be taken courses to take marketing strategies, strategy, strategic brand management, product marketing fundamentals. These all will help to build strategic marketing. A data measurement and attribution, so marketing analytics, attribution modeling and measurement, all these can be the one. So now it's suggesting all the kinds of trainings which Tun Moy can do to upskill himself to bridge the gap. Now, what Turmo must do in the next 12 months not learn, refrain from his. So now there's some internal changes which Tnoi should do with respect to his current role. Rather than positioning it as a Google Ads expert, he can position it as a marketing impact owner. Rewrite resume and LinkedIn. Okay, build some two, three portfolios, add marketing exposure to his portfolio. Okay? So this way, we are able to you can see the steps are given very clearly, specifically what are things to be added to the current profile in order to match up to the new job role the person is seeking. So 12 months development roadmap. So what needs to be done in the first three months, completing the marketing strategy and analytics courses. What has to be done in the next four to six months, seven to nine, and so on and so and so forth. So readiness assessment, which can also be done here, which is basically channel expertise is good analytics, which requires refraining from which needs proof. All these things needs to be taken care of as a readiness assessment, which we have to do. And finally, the reality statement doesn't need more PPC depth. What it requires is. These are things which are needed and which you can bring in in the next 12 months to be eligible for a digital marketing manager role at Google. So you see, with the help of the AI tool, it gives the complete framework. It gives a complete framework from start to end what all things the person needs to do to grow to the next role which he or she is driven for. So that is how you can make use of this AI tool to come up with a PDP plan, a professional development plan for any of the employees in your company within your HR scope of work to build these kind of PDP plans for employees who are looking to move to different roles which they want to pursue. 21. Forecasting Employee Growth and Readiness for Human Capital Planning: Hi, yes. Welcome to this session. In this session, we wanted to see that the PDP plan which we created, right, for the employee, we can also the great part of using AI tool can be that we can also make tweaks to it while the plan is going on. Okay? While the plan is active and it's been executed, we can make some changes in the sense that what updates have come in, what all courses the employee has already done, and with that, how the trajectory changes. So that also will be possible with the help of AI tool, wherein you can just update the tool about what a new things has happened so far, and that is going to give us a new forecasting for the PDP plan for the employee. So let's see this in practicality, how it is going to be. So let's say we are giving an update to the AI tool regarding Tan Mois progress wherein we are telling the tool that he has completed certain courses. So with that, how does the actual development tretory changes? So we just want to know that. So we're saying that project when he will be ready and show the updated timeline for the role based on his actual development trajectory. So he's looking for a digital marketing role at Google. So how does that rejectory changes? That is what we want to understand. So we have also telling that these are three courses which he has completed, in the last year, okay, prompt engineering for hat GPT advanced, prompt engineering for everyone, and open AIs GPTs creating your own custom AI assistant. So with this coming into picture, the changes are being shown. Original estimate was 12 months, okay, which was given. But the revised estimate based on actual dejectory is 14 to 16 months. AI courses are additive, so not substitute. Okay, because he's looking for a digital marketing manager role, which is in which, again, AI courses will be additional. So it's going to give him more time, so his trajectory will increase in this particular case. So with that, it is going to tell about how it's going to help is that the AI courses are going to be supporting skills which he can certainly use. It will help him in course strategy and measurement building, specifically, which he can do. He can complete the marketing strategy courses marketing analytics courses, which is going to be more aligned with his new role which he's pursuing. Same way, experimentation case proof of complete experimentation, growth marketing courses, all these are going to help. So now the tool is giving up phase by phase, execution which we need to do with respect to what all things need to be covered so that he's able to reach he becomes eligible for that digital marketing manager role within the next 14 to 16 months. So this is also which we can practically do with any of the employees, once they start taking those courses, once they are executing all those steps which are needed when they're pursuing a new role within the organization. 22. Accessible Explanations: Get the Key Ideas Right Now: Hi, guys. Welcome to the sessions. In this session, we'll talk about how we can make use of generative AI in HR, specifically for getting explanations in the way which we want. A lot of times what happens is the information which we have and we need to reach out to a lot of people in the company with respect to the employees. We need to get the information, we need to understand the information in a specific manner and that might not be the case. Becomes really difficult for us to understand a particular, let's say, a job profile, which has specific technical terms or jargons being used and becomes extremely difficult for us as HR employees to go ahead and understand that and based on which take the necessary steps. So in such cases, we can make use of the generative AI to transform that kind of information into accessible information or explanation for us, explanations which we can understand. You can transform that and then you can go ahead and do that. So let's see a practical example of what we are trying to say here. Let's say we are here to interview a specific candidate and we can upload the resume of that particular candidate and we can ask the genitive AI to go ahead and have a look at the document and make it explain the job role, the expertise of that person in understandable manner. So here is the prom which we can give where we are saying that we're going to be interviewing the megas for digital marketing manager role at Google, I'm doing an initial interview, but I am in HR, and I'm not an expert in the field. Don't use any jargons or technical terms, explain things clearly using analogies and concrete examples that don't require domain knowledge. I'm setting the expectations. I'm setting that how I want the explanation to be given to me. This is the power of AI where you can ask it to provide the output in a certain manner. Now, this can be and now this can go anywhere. You can ask it to provide information with domain knowledge, without domain knowledge. You can ask to give the explanation with, let's say, keeping in mind that the audience is going to be CEOs. So you can give any type of expectation and based on which the AI tool will customize it and provide the output. I'm also saying that I want you to help me get to know the candidate and most important accomplishments that I can discuss with him. What are some important contributions that they have made and that I can discuss with them? Write two paragraphs 2 to three paragraphs of the narrative to help me get to know the candidate. So now what is happening is the AI tool is transforming, changing the whole resume, which might have a lot of technical terms and jargons into simple understandable language which I'm able to understand. This is what we mean by accessible explanations, which we can have. And now I can understand better what the resume has to tell me and also the specific things we asked for, it has pointed out. Now, apart from this, I can also go ahead and tell the AI tool that now explain to me in one to two paragraphs, what are the most important parts of this role and what skills or qualities is the team looking for? Explain in non technical terms. What is the expectation? What is the job expectation that also I want to clearly understand, possibly I can share that with the candidate in the interview. So you can see what is happening is, uh this is one of the examples where what you can do is any type of technical terms which comes across, maybe a specific document which has been shared with you, from the senior management related to certain changes in HR policies, which is too much technical and you are not able to understand complete, you can upload here on the generative AI tool and ask it to present it in a understandable manner. So that is the power of the AI tool, and that is how you can make your work much more simpler and easier to understand, uh with the help of how it can transform any type of information. 23. Question Generator Pattern: Hi, guys. Welcome to this session. In this session, we'll see how we can make use of the generative AI to generate questions for us which we need while we are working in HR specifically. Let's say in terms of questioning or interviewing a specific candidate. So in this particular case, we're taking a situation wherein we are not the expert of that domain, but we want to ask certain questions which might be useful to assess somewhat some level of understanding of how the candidate is and whether he might be a suitable role profile for the role which we are looking. So from that aspect, we want to use the generative AI to ask those specific important questions and see how well it is able to assess the candidate for us. So we can use the generative AI to generate those questions which we will be asking. So let's take an example of this particular resume, which is what Tarmgas and we can go ahead and give the particular query over here, okay, wherein we are asking specifically to help me create one or two questions to assess the candidate's level of excitement, specifically. What we are going to do is we're going to ask this particular question, which is I'm not an expert in the domain, but I want some simple questions that I could ask to better determine his suitability for the role in the attached job description. I don't know the technical details, but I want some probing questions that I could ask. I will provide my understanding of the answers back to the team looking to hire to help them decide if they want to schedule an interview with him. So this is the initial phase of interviewing which we are doing from the HR department. Okay? So now it has come up with those questions. Can you tell me about the time you helped a business or client grow, even though they were unsure or hesitant at first, right? So this does not require any technical jargons. There is no specific things about the profile. We are generally asking to understand the candidate's mindset. How do you usually explain complex ideas to someone who has no background in marketing, right? It also gives you what this reveals and what we are expecting from the candidate, what we want to listen from the person. So these are the questions which we can get. How do you decide what to focus on when you have many responsibilities at the same time? Listen, clear prioritization logic, structured thinking, comfort with changing priorities. Now, once we have this, we can also ask AI tool to help us with other questions, which can we look at the candidate's resume, now generate two to three questions that are more probing and we'll help assess if this candidate has the most important skills, one to two skills or not. Okay. So now it is going to look at the resume, and from there it is going to assess. The first question it comes up with is you have spent a lot of time training and coaching others. How do you know your training actually worked? Some people once people were back on the job, right? So this is a good question to connect back to the impact, the training impact, which we are trying to assess over here. Clear ways he measured success, examples of adjusting his approach, okay, ownership of outcomes. Tell me about a time you had to push for a change that help the business even when people were uncomfortable or resistant. So these are the type of questions now it has come up with. Now, other than this, you can go a little deeper to ask for other questions which can help me create one to two questions to assess this candidate's level of excitement to work at Google, and to determine his motivation for changing jobs. Okay? So now we want to have questions around these two points. So you have worked with Google in different ways as an employee trainer on Google programs and someone who teaches others about Google products. What specifically is drawing you back to Google Now, right? So it connects back to that point and why this role at this point in your career. If you stayed in your current path for the next two years, what would be what would be missing for you and how do you believe Google fills that gap, right? So this is how we can get more intriguing questions generated with the help of AI for an interview per se. And lastly, let's change this whole thing, and let's say that let's think of one last very challenging question that a known expert like me could ask that would help assess this candidate's knowledge of the area and help determine how well he thinks on his feet or collaborate with others. Okay. So now we are giving this particular thing. Imagine a business partner is unhappy and says, I've been working with Google for months and I don't feel it's helping my business. You're not sure yet whether the issue is the product expectations or how they're using it. Walk me through. What would you do in the first conversation and who else, if anyone, you would involve? Right? So now we are asking for apart from the job role, we're asking much more deeper questions specifically, understanding the understanding of the candidate of the role and what all things he can bring on the table. So you see, this is how we can make use of the genertiveVI for generating questions for us related to our HR, maybe useful a lot in the initial interviews which we take for the candidates. 24. Standardization Pattern: Hi, guys. Welcome to this session. In this session, we wanted to see how we can make use of generative BI with respect to standardizing certain processes. So once you are working with HR and you're interviewing various people, various candidates, you would like to have a standardized process or metrics based on which you want to evaluate your candidates and then take a judgment yourself whether you want to proceed further with that candidate or not. So for that as well, we can use generative AI to build this framework for us. Let's try to understand what we are trying to do here. So let's say we have a specific job description for which we are hiring right now and we want to create a standardized framework or evaluation comparison sheet, which will help us put in all the details of the candidates and then do comparison between them to understand to whom we want to go forward. So this is the prompt which we are giving. I am in HR and helping a team hire for attached position. This is the position. I want to create a standardized table that can capture key information or skills from applicants. I want to standardize and highlight key information needed for people to compare those candidates based on the job criteria. I want the criteria in columns and the candidates will be in the rows. We need a reasonable number of columns that capture the most important candidate skills, qualities or experience. So now we have the structure in place. So here we can add the candidates names and then these are the criterias it has created. You can see relevant experience, business impact examples, all those coming from, as a structure which we are building, ability to explain complex ideas. These are the criterias on which we are going to assess our candidates and then compare them eventually to know to whom to go ahead with. So now it has created this. Now what we're going to do is we're going to apply this on a specific candidate. Okay? So we have uploaded the candidate's resume. We have uploaded the comparison sheet as well, and now we give it a prompt where we say that read the attachment, resume, and the candidate information sheet for each column include direct quotations. Okay? We don't want explanations or generic information. We want exact quotations from the resume that would help be helpful to the person evaluating the candidate. Those can be added here. We don't want to judge the candidate, so we are not asking the AI tool to come to a conclusion on the candidate whether to go ahead with them or not. That decision is being taken by us. We just want it to help us with the comparison to be done, standardization of metrics to be done. All that is being done by the AI. We don't want to judge the candidates. We want to standardize the presentation of information to help the human evaluators. If you can't find anything relevant, then we're also explicitly saying that leave that particular metric black. Create an updated version of the sheet with the information in. So now you can see it has gone ahead and provided that information, so candidate name, relevant experience coming from the resume, Business impact, given ability to explain complex ideas, simply, okay, data and decision making mindset, cross team collaboration, all this coming from data points coming from the resume itself, ownership and accountability, comfort with ambiguity and change. Communication and influence. Okay. All these being provided over here, awards and recognitions, motivation for Google role, all that is being provided here. This is how we can go ahead and standardize the processes which we have in HR specifically and then get the output in a much faster manner and effective manner as well. This is one of the examples which you can use very easily in HR to hire people. I hope this makes sense. I hope you understand now how we are making use of the standardization pattern in generative AI for our HR work. 25. Introduction to Generative AI in Talent Acquisition: Hi, guys. Welcome to this session. In this session, we'll see, we'll talk about how generative AI can be useful in talent acquisition processes as well. So if you look at the practical applications of generative AI and talent acquistion onboarding, can be multiple. Like the first is with the help of AI, you can have a personalized candidate communication, which is basically you can generate tailored job descriptions for the candidates, emails which can be sent to them, customized and interview questions which are much more relatable to the profile which you're hiring for and the candidate the role which you're hiring for and the profile which you're interviewing. There can also be automated screening and short listing, which you can do with genitive AI, where you can automatically with AI, analyze the resume, understand the strengths of the resume, assess the candidates fit with the profile. That process you can set up with AI. Also, you can automate the initial onboarding screening process. Which candidate to select, which one not to select, considering the strengths and weaknesses, based on which the first screening process can also be automated with the help of AI. There is also going to be interactive onboarding experiences which you can create with generative were where you can create engaging onboarding material like personalized welcome messages, interactive training modules, generative AI powered chat boards, which can be a good user experience for the new candidates. Also, uh, there can be content generation which you can do for training purposes. So while they are getting trained, you can create quizzes, you can create summaries, scripts for video training videos which you might need catering to different learning styles. These are all the benefits which you can see, which you can have, and you can build with the help of generative AI in TA or talent acquisition and onboarding. In addition to this, bias mitigation, which can also be done here, which is you can utilize genetic PI to identify and mitigate potential biases in job descriptions. So this can be really a fair process which you can build for candidate selection, performance evaluations, you promoting fair and inclusive hiring process. So, there can be multiple different use cases of generative VI with respect to talent acquisition and onboarding processes. 26. Generative AI Use Cases in Talent Acquisition: Hi, guys. Welcome to this session. So in this session, we'll see some use cases of generative VI in talent acquisition. So what we understand here is there are multiple things which you can do in talent acquisition with respect to genetive VI. The first is going to be automated interaction. So here, you can build automated communication which is needed initially when you are communicating with the new candidates and also from job descriptions which you can create, you can create you can send out automated follow up emails, improving the engagement and saving a lot of human time. Then you can look at content generation as well and management, which basically generates where the AI tool can be used to generate tailored job postings, interview questions which we can create through the AI tool and the onboarding material as well needed for each of the roles. There is also going to be inside generation and personalization which you can do now. You can analyze the candidates profile to quickly identify the top candidates with precision. This is going to take much lesser time comparatively than manually going through each and every profile. You can also do agent assistance and workflow automation, which is basically streamlining, scheduling of interviews, task management, routine recruitment workflows which are needed. Those can be generated through AI, which allows your teams to focus on bigger strategy discussions. Now, with respect to this, you can also identify a lot of generative AI use cases in talent acquisition and onboarding, which can be, first of all, looking at identifying the manual and time consuming processes. What are the main manual time consuming processes? You can identify those tasks that involve a lot of human manual effort. Like resume screening, interview scheduling, onboarding, paperwork, all these can be given to the AI tool to do and which saves us a lot of time. Second, what we can do is we can look at highlighting areas which are prone to human errors. Look for processes which can be human errors such as data entry jobs of compliance checks, candidate communication. All these can have a lot of human errors, which we can eradicate by assigning it to a generative AI tool. Then there is uncovering opportunities for deeper insights, which is basically considering areas where you want to gain more understanding and candidate fit or onboarding effectiveness, diversity and inclusion. For these specific things, you can take special efforts by taking help of the AI tool to generate that content for us. Is also going to be exploring available AI solutions. So you can research for generative AI tools designed specifically for TA and onboarding, including generative AI powered ATS, right? The portal where you can manage all the profiles. You can create chat bots, content generators, which can be built with AI tools, which can generate various material needed for TA and onboarding. And then analytics platforms as well, which can help you understand the which profiles to focus on. The last thing which we should be doing is when you start applying these generative AI tools on TA and onboarding is starting with the pilot projects where you begin by implementing it the generative AI in smaller areas to test its effectiveness. For example, automating only the resume screening or piloting a generative AI chat board or personalizing the onboarding materials which you want to create. So all that, you can start as a testing phase, experimenting phase. And then once you see the output of it, you can build similar tools, similar processes for the other parts of the TA and onboarding. I hope this makes sense. I hope you understand now how we are trying to use the generative AI tools to incorporate in our TA and onwarding processes. 27. Mastering Promot Engineering and Develop a JD Creator Part 1: Hi, guys. Welcome to this sessions. In this session, we'll see how we can master prompt engineering and develop a GD creator specifically, how we can generate a specific resource for TA and onboarding with the help of effective prompt engineering. So prompt engineering is going to be most effective part of the EI usage wherein we need to create and provide effective prompts, which are key to extracting accurate outcomes from genetive AI and LLM tools. So we have to make sure what kind of prompt are we giving for the desired output which you're looking for. So the more specific you are going to be in your prompt, the more effective responses you're going to get from these tools. The prompt acts as basically instructions which are guiding you're giving to the AI tool to understand your needs and based on which it is going to provide its outputs. Now, a well written prompt are going to be essential when you are trying to make your T and onboarding process a success. So we have to customize our prompts around our T and onboarding task and give it to the AI tools to provide the right output. If you look at a problem area, let's say, a scenario wherein the onboarding process has been very overwhelming for the new hires. So in such a case, the objective which you have in place, the problem which you have in place is to streamline the onboarding process for the new hires with the help of genitivI. So how you're going to do that is you can break that into four parts. The first part is going to be instruction where you're going to clearly state the problem, what you want the AI tool to do. So in this case, the instruction can be to analyze the current onboarding process, the materials being used, and suggest improvements based on new hire feedback which you have received. Once you get this from the AI tool, then you set the context. We set the context, provide the EI with the background information that the new hires were feeling overwhelmed with the onboarding process and they were struggling with the navigation of the onboarding portal and they don't have a clear understanding of their roles and responsibilities. You're giving the context. Once you provide all of this to the AI tool, then the question which you can give, which is where you directly ask the AI tool and what we want it to answer. We're asking how the onboarding process, the material can be improved to address the issues new hires are facing. Now with that, the AI tool can provide you the output. Okay? So this is where we guide the AI on how we want the answer to be structured. What kind of output are you looking for? What is the expectation? So we ask for specific recommendations like simplifying content, how we can simplify the content, how we can improve portal navigation, clarifying roles, elements of the TA process can be streamlined and personalizing the overall experience for different routes. So this is how you can structure an effective AI proms to solve a specific use case scenario which you're facing in TN onboarding. There are certain best practices as well, which you can keep in mind while crafting these effective AI prompts. First is, we want to keep our prompts as simple and clear as possible to ensure there is no ambiguity, and it is easily understandable the tool for the tool. We need to be very specific and direct with the instructions to get accurate results. We spoke about this earlier as well. We also need to provide relevant information, context to help the AI tool understand the current scenario, the task, which it needs to solve and generate useful responses. Uh, we also need to define the desired output clearly to match our expectations. What is the kind of output you are looking for? So it gives the e tool more context around how it is going to frame the responses and provide that to us. Then we can test and refine the prompts as well going forward regularly to improve the clarity and effectiveness, to get better results, to get more effective practical outputs responses which we can actually put into action. Lastly, we can also make sure to avoid any ambiguity or confusion by specifying what should and what shouldn't be included in the response. This the AI tool is very clear about what kind of response it has to generate for us. I hope this makes sense. I hope you understand now how we are going to make use of the AA tools from engineering to help us in our TA and onboarding process. 28. Mastering Promot Engineering and Develop a JD Creator Part 2: Hi, guys. Welcome to this session. In this session, we'll see how we can make use of the AI tool like hat GBT to develop a job description, creator specifically. Create a job description for a specific kind of role. Let's start with the interface first. This is how the hat GBT looks like now. We are on a Chat GBT 5.5 0.2 version right now, which is what it is right at this moment. And this is if you look at the plan, we are on a go plan. We are going to go plan right now, which is a By plan. You can also do this on a free plan as well where the output can be a little similar, but the go plan gives you additional benefits as well as you can see mentioned over here, which we get to see. I would suggest it would be a good idea that you can take up a go plan and that gives you much better outputs. Apart from this, if you look at the settings, there are certain things which I just wanted to mention. For example, you can go to data controls where you can switch on the improve the model for everyone option. So this basically what it does is, when you switch it on, then HAGEPT is going to take into consideration your previous conversations, your content, specifically to train its models to give you more customized solutions and results. So totally your choice if you want to do that, but if you feel it's privacy issues, then you might go ahead and switch it offers as well. You can do that. Other than this, you also have the apps option where you can connect your tools, other tools like Google Drive or any other tools, which you can connect to hatGPT so that the transition usage can be much more seamless and it becomes easier for you to work with this particular platform. Now, let's go back to the prompt and look at what we are here for. So let's say we want to create a job description for a specific role. So let's look at two options. The first option is where we can give it a specific prompt, something like this, where we say, create a job description for senior data scientist. This is a straightforward simple prompt which we are giving. And with that, ChatBT can start doing that job of providing us with a complete job description, looking at the job title, location, employment type, role, overview. So all these are being given. Required qualification, preferred qualification, what we offer, Okay, so all this is provided. Now, this is absolutely fine. However, you have to obviously in a real life scenario, you will have to check this with your current requirement when you are planning to post a job description for a specific role, whether it aligns with that or not. So you will have to do that manual mapping of what Ta JBT is producing versus what you really require in your HR work, right? So the other option which you can do is you can give it a separate different prompt where the prompt can be a little bit more specific. You give a very structured prompt exactly what is your requirement. And based on which then Chat GPT gives the output. So let's have a look at that also. So this is coming from the point which we had discussed earlier, which is a framework of instructions, context, question, and output, right? So in instructions, we give the clear instruction that you are to create a comprehensive job description based on the provided role, requirements and context, ensure it appeals to the qualified candidates and aligns with company standards. Then you give the context, his role is this department, team size, company culture, skill required. You mentioned all of that. Question, based on the context above, draft job description, we have to create a job description highlighting the primary responsibilities, required qualifications, preferred skills, and benefits associated with this position. And then finally the output. We want the output in this particular manner where there will be job title, company overview, role summary, key responsibilities, preferred skills, and benefits mentioned. Now this is much more informed detailed information which we are providing Chat GBT, so it gets a little bit more informed about what is expected out of it, and possibly it will give you a much better output versus the previous one. So let's have a look at this. So we're going to give that particular prompt. We have added the prompt over here. So now it is going to create that. So you can see it's creating a job title company overview, role summary, key responsibilities. Which we want from it. It's going to generate all the information based on the context which we have given right here. Now it has given us the key responsibility required qualifications, preferred skills, benefits, all of that being provided. This is getting a little bit more specific like experience working in cloud based environments, Abu as GCP, Azure. This looks much more informative than the previous one if you do a comparison et. This is how we can make use of Chat JPT for job description creations, which can be a part of our HR. Okay. 29. Create a Resume Screening GPT: Hi, guys. Welcome to this session. In this session, we'll see how we can create a custom GPT specifically for resume screening. Let's try to understand what is a custom GPT. A custom GPT is primarily, you can imagine it like a personal assistant, which you have built on the GPT platform. Now, this is going to be a specific kind of a GPT customized for your solutions, your problems in your business. Okay? So here, you can give it certain instructions, conversation starters, knowledge base, capabilities, actions, which you give. And with that, you build out the structure of this custom GPT, which is customized to provide solutions catering to your specific needs. So in instructions, this is where you're going to tell the AI how you want it to behave, how to give the responses in a formal, casual or focused manner. Conversation starters are going to be examples that show that the AI, the kind of conversations it will be handling. Okay? Knowledge base is this is where you're going to go ahead and feed the AI with specific information it needs like company policies or product details, FAQs, going to provide all of that. In capabilities, these are going to be some extra features that you can use with your GPT, like web browsing, if you want to do so that will be available, generating images with Dali, all that will be possible. And then comes actions. In actions is where you allow to connect your GPT to specific services like APIs, adding documents, all that can be done. With all of this, we build a custom GPT, and we give all the instructions of what problem are we trying to solve. We give the background information as well to it, and then it starts providing us results catering to our specific requirements. Let's see this in practice how we are going to create this for resume screening. Our intent is that we want to build a custom GPT for resume screening where we want it to screen a specific resume for a specific job description. We have a specific job description in our HR. We have rolled out a particular opening and there is a resume which we are receiving and we need to see whether the resume matches that job description or not. That is what the job of the custom GBT would be. Let's try to build this out. For building custom GPTs specifically, we're going to come here and we can explore GPTs and this is where we can create. We can create new GPTs ourselves, and this is the GPT store where you can see there are a lot of different GPT, custom GBTs created by other people. Similarly, you can also build your own and you can create a new one where you have the option to create from here. So here you can give all the details. You give the details of what kind of a GPT you want to create, and it will start building it out the model for you. Okay? So let's say this is the prompt which we want to give. So create a JBT that takes in a CV and a JD and provides a critical evaluation of whether or not the CV is suitable for the position. Output a list of points in favor or against the key requirements from the JT and provide evidence. Provide a score of zero to ten against key requirements, give a final recommendation on whether to proceed with the resume or not. Additionally, provide a comment on what kind of position this candidate will be best suited for. Okay, so this is what we want. So we give that to hat GPT, and now it will start building the custom GPT for us. It is going to ask us to recommend or suggest certain name for this custom GPT. Okay? So resume Fit. Yes, it's giving it a certain name. And now it's generating a profile picture for it as well. You can see on the right hand side, this is where the custom GPT is getting created. You can see some example prompts which it is automatically generating. So it's created the profile picture as well. Okay. If you want, you can customize it as well, and let's look at the configurre option. So here you can see the name of the custom GPT. If you want to change that yourself, you can do that. If you want to give a one line description of the GD of the custom GPT, you can do that also. And this is the information we have provided. Now, these are the conversation starters. This is what we're talking about. Conversation starters are going to be the different types of proms which it will deal with. So those are given over here, and this is what we exactly want, right? Evaluate this CV against the attached JD. So once you have this, you can also upload certain files just giving additional resources to the custom GPT to be well equipped to provide us better results. So you can do that also. And then the capabilities which you want to switch it on, you want it to be able to do web search, use Canvas, okay, use image generation. So all that if you want, you can switch them on as well, and then you create. Once you create this, this particular custom GPT will be in action, active, and then we can go ahead and use that. So right now, let's say, these are the options that will give you. So if you want to uh, keep it with yourself private. For now, you can do that or anyone with the link or put it on the GPT store. So let's say we are doing it only for me for now. Once you have the GPT created, then our job would be that as the purpose of this particular customer GPT was to assess a particular JD for a particular role. So here, you can now simply this is our JD custom GPT created, so we can upload let's say we'll try to upload one JD, and we also give a job description. And we say, please evaluate This is a requirement of ours. Now it is going to look at the documents. And you can see, for the key requirement mapping scoring, it's giving us the score, core functionalities, budget planning, all of that is giving O score is 7.1 out of ten. So now based on this, we can understand, and it gives the strengths and gaps as well to assess. So possibly you can ask if you have a particular benchmark. Let's say you want to uh call in all the candidates for interviews who have scored more than five. Then once you interview this particular person, you can dig more and ask questions around the gaps and risks. Higher recommendation. Do not proceed. For now, it is saying do not proceed for a P Digi marketing role. Proceed if role is adjusted, alternative roles are considered. Better alternatives are also being given out here. So this is how we are going to create a custom GPT guys, which can be useful in HR works. This is one of the examples of resume screening which you can create as a custom GPT and make use of it in your day to day HR work. 30. Automate Resume Screening using Gemini: Hi, guys. Welcome to the sessions. In this session, we'll see how we can go ahead and do resume screening as well with the help of AI tool like Google Gemini. Google Gemini is another AI tool created by Google, similar like Open AIHatGBT which you can use for providing getting outputs based on the prompts you give it. So let's have a look at this tool. So this is the platform guys, which is Google Gemini, which you can certainly use over here, and it has a free and paid plan as well. So you can absolutely go ahead and have a look at that also so which you can make use of. This is what we're going to use. What we're going to do here is ideally, we are going to go to the settings of it wherein we are going to make sure we are actually using it for different Google products, which is going to be Google Drive. We're going to use the Google Gemini linked to the other Google products like Google Drive, Google Excel, Excel spreadsheets specifically and see how we can use that for automating our resume screening process. So this is our Google spreadsheet which we are using, right? So here we are going to use the particular Google Gemini tool. So for that, what we have to do is we can go ahead and activate it from here, which says Ask Gemini. And now we can start the whole process of giving it specific prompts related to our resume screening process. So let's say this is the first prom which we want to give, which we wanted to create a table with a list of five sample candidates with the following columns, which will be names, emails, available date for interview, and available. So now we have given that to the Gemini tool and it's going to create the table for us. You can see the table has been created, which you can absolutely go ahead and edit as per your details. So you can do that, as well. So we can say, let's insert this here itself. So we have the data over here. Now what we want to do is against these specific. So you're going to update all the details, the names, emails of your interviewers, interviewees. And now against them, we will have to add their specific, you can say the interviewers who are going to be there. Okay? So let's give it this prompt. Now we're going to give the prompt over here where we have two interviewers. Let's say Tamdas available on, uh, tenth February, there is 11th February 13, February, and Greg is available for 12 and, um, there is on tenth, 11th and 13th, 12 and ten. This is how the available, the interviewers are available for it as well. Now we want the we just want to put the interviewers against the interviewee so that we know which are the people who are going to be there. So now we have that over here. So interviewers are assigned. So let's put it as well. So this we have in place. Okay. Now, once we have this, so we have the details of it, we want to send out emails to them inviting them for the interview, right? So for that, we have a specific prompt which we want to give where we say that now generate individual interview mailers for each of the candidates, appropriately greet them and thank them for taking the interest in the role of a software developer, and mention interview date and timing, interview name, standard, interview etiquette to follow and wishing them all the best. Provide a fictious Google team link as well to join for the interim. So this is a email, which is going to go out to each of them, inviting them for the interm. So now the Gemini tool will go out and generate those emails as well for us. As you can see, the first is for Alex Johnson, which is correct, Alex Johnson. Okay? The email has been created, drafted, which is for tenth of February. Data is correct. Okay, interviewer is going to be thermidas the Google Meet Link provided. Standard interview etiquette provided. Okay, same goes for Maria, the next person, okay? And interviewer is Greg Smith. Okay, the details are absolutely correct. You can see now all the emails have been composed. Okay. All you have to do is now send it out from your official email to all of them. So this is how we can make use of the Google Gemini tool to do a lot of HR work like resume screening. Okay? You can also do candidate evaluation, okay, job description, creation. A lot of these stuff can also be done. 31. Create Candidate Evaluation GPT: Hi, guys. Welcome to this session. In this session, we'll see how we can create a candidate evaluation GPT as well on GPT specifically, a custom GPT, which is going to basically going to evaluate candidates. So for this, what we require are three major things. One is going to be a case where we want to create we need to have a document containing the JD, the job description of the job profile, which we're looking for evaluation parameters based on which we are going to judge. And the answers given by the interview VE, by the candidate specifically, which is what the GPT is going to assess. So the GPT is going to assess based on these and then score them on a scale of let's say zero to five. So we have created a specific prompt for that also. So this is the prompt we're going to use to create that custom GPT. Let's have a look at this. What we're going to do is we're going to go to explore GPT, and we're going to create a new GPT altogether, and we're going to give this prompt. The prom clearly says, create a GPT that takes in a document containing a JD, evaluation parameters and a set of questions answered by the candidate. The GPT has to provide a critical evaluation of the candidate and also score them on each of the evaluation aspects on a scale of 025. This is what we want to make. And based on this, then we will have the scoring done by the GBT, whether they are fit to move to the next level, next rounds of interviews or not. What we're going to see is we're going to take two different scenarios. One scenario can be where the candidate is given the right appropriate answers needed. We can see, yes. We can just reply to this. Okay. And in the second scenario, we will see a candidate whose answers have not been up to the mark. So we want to see whether the custom GPT is able to make that difference and score them accordingly. So let's have a look at this. Right now, we are just creating the custom GPT and then once it is active, we can upload all the information. So just imagine having this GPT already with you. How this is going to be really useful and saves a lot of time is you might be having a lot of people getting interviewed. Now you quickly want to score them whether the screening process, the evaluation process is going on. So now we have the custom GPT created. So the idea is that with the help of this, you can save a lot of time of candidate evaluation process. You just need to upload their answers, and you will have the valuation parameters already fitted in. So with that, you can simply go ahead and create it yourself and you can upload the documents one after the other in the custom GPT, and it will score you. You can get the scoring proper scoring can be done, and based on which you can then decide the next steps, which are the candidates which are moving forward to the next rounds and which are not. That's the idea. Now we have the GPT created. So now we have the candidate evaluation, GPT created for us. So now let's upload the document. So I have already created the document over here. So the first one is going to be this one, which contains all the information. So let's have a look at it as well. So now we have given the information, so let's have a look at it as well here. I just wanted to show you what information we are uploading. This is the role of software developer, job description is provided. Then we have given the interview goals, evaluation criteria also mentioned, and then the questions answered by the candidate. This is the document which we have here, and this is what we are uploading in hat GPT in the custom GPT. Es. So now let's see it is going to look at the document. So now it's going to evaluate based on the parameters suggested. So technical competency, four out of five, practical experience, 3.5 out of five, problem solving analytical thinking, 4.5. It's measuring based on that. The overall average score has been given us 4.1. So we can see this is being provided right here. Now let's look at another example. And we're going to give the same prompt. Demonstrates basic awareness of software development concept, but his interview responses reveals significant gaps. You can see technical knowledge is 2.5. Practical application is also 2.5. Problem solving ability is very low, 1.5 out of five, communication skills, learning and growth potential, 2.5. The overall score is now coming to 2.1 out of five. Like this, you can use the custom GPT primarily for candidate evaluation and quickly the tool can evaluate based on your parameters, your interview evaluation parameters and tell us whether the candidate is good fit to move to the next round or. And 32. Develop a BGV Automation GPT: Hi, guys. Welcome to this session. In this session, we'll see how we can make a custom GPT, which is background verification automation. Here, what we are trying to do is we are trying to verify the details given by the candidate with what is mentioned on their CV. Whatever certifications they have done and does that show up in the same manner in the CV or not? That is what we want to verify with the help of custom GPT. So let's have a look at this. We're going to create a custom GPT by going to explore GPTs and create. So this is where we're going to give it a prom. So let's say this is the prompt which we want to give. Create a GPT that takes in a CV and testimonial documents. Please validate whether the CV and the testimonials attached, uh, tally. If there is an anomaly, then please mention so we're going to use this and create a custom GPT right now. So this will make our work much more faster because then with the help of this custom GPT, you just need to upload the verification documents provided by the candidate and their CV, and it can tell us a yes or a no on that. So you can see it automatically picks up the initial prompts which you can use. So giving it a name as well, which is CV testimonial validator and it's also generating a profile image of it. So once we have the custom GPT in place, we will upload all the details. We're going to upload their CV. We're going to also upload their document. Okay. Let's take a look at this. So this is first of all, the document, which is the CV, which we are uploading for Greg Smith. Let's say, that's the candidate we are verifying here, and then we're going to upload the other documents. The certificates which we're uploading right now. So now we uploaded and we're going to just ask to please evaluate going to look at the CV, it's going to look at the certificates and then verify and check whether that is the case or not. It says that both PMP and CI essay are there mentioned, confirmed and consistent. So we can verify that. It is completely verified that the certifications documentation which the candidate provided is clearly mentioned in the CV as well. I hope this makes sense. This is how we're going to make use of the custom GPT for any type of background verification, automation which we want to do for our HR. 33. Develop an Onboarding Chatbot: Hi, guys. Welcome to this session. So in this session, we'll see how we can make use of Chat GPT to develop a onboarding chat board for our new joiners. So new joiners, once their interview is cleared and they're about to join, they can and they will have a lot of questions related to company policies, dress code being followed, okay, the holidays, information, weekends off, all those information which they would require to know number of hours working overtime, knowledge about that leaves information. So for those, they will have a lot of queries which they would like to get answered for. Now, providing answers to all of these queries manually on a daily basis can be a very overwhelming and time consuming for the HRT. Other than that, what we can do is we can build a custom GPT which can help manage all of this. Let's see how we can do that. Once you are on CTA GPT, we can start building a new GPT altogether, where we're going to create a specific GPT, which is going to cater to this particular scenario. We're going to create this particular one. Where we say that create a GPT that has the knowledge base of the company's HR policies. It takes questions and answers them purely from the policy document provided. This GPT is intended to help new joiners in clarifying various policies related questions. Also, we also mentioned that please respond as I do not know when the answer to the question is not available in the given knowledge base. Here, we will also have to attach the knowledge base or the policy document in this custom GPT. We're going to take this at this over here and we're going to add the custom GPT as well. We can also upload the files right here. So we're going to put the HR policies. We have uploaded the document, as you can see, and it is going to create the name as well, it has created it has created the document HR policy has been uploaded here, and then we can create the whole custom GPT. So now based on which you can test out the GPT as well. Um, So let's have a look at that. So this is going to be really useful because it's going to save a lot of time because the new joiners will be very inquisitive and will have a lot of questions in their mind, which they can easily go ahead and get answered for from this custom GPT. So this is the custom GPT which you can copy, and now you can share it with all the new joiners in the team. So let's test this out as well. So let's say we want to know about the leave policy. So just understand that this particular custom GPT has the policy document at the back end as well. So whatever answer it is going to give is going to give based on the document it has been attached to. So it's going to look at the document and pick up all the information from there. Let's say we're asking more information, please explain extraordinary leave. So now it's giving us more information, looking at the policy document. This is how we can make use of the chat GPT to create a custom GPT specifically catering to. It's works like a chat board which is going to answer all the queries related to on boarding. I hope this makes sense. I hope people to understand now how we are using the AI tool to, uh simplify our HR related processes and work which we have and implement automation as much as we can possible in our day to day work. 34. Best Practices for AI in Talent Acquisition: Hi, guys. Welcome to this sessions. In this session, we'll discuss about the best practices which we can keep in mind for AI in talent acquistion specifically. So the first is kind of identifying the key use cases. So we need to focus on tasks where AI can make the highest impact, like resume screening, interview, scheduling or managing on voting documents. Automating these kind of tasks can free up a lot of time and help the teams to discuss about higher strategy areas. Also, you can look at maintaining human oversight. So while AI is being used on a regular basis for doing the heavy lift humans can remain involved, especially for complex decisions to be made, final decisions to be made like final candidate selections. This way, you're adding the thoughtful personal touch to the whole process. Training with relevant data as well. So using specific data from organization like hiring trends or job descriptions will help to train the AI in the right manner. It ensures that it provides outputs relevant to our business, to our industry, specifically, catering to our needs. Also implementing continuous learning. So regularly updating the AI with new data and feedback can help it to adapt to the changes in the job market conditions, policies, practices, and keeping it much more relevant and accurate. Also, we need to ensure that there is clear communication. So transparency really matters when we are using AI tools. So everyone knows when AI is involved, so they can review or adjust the results, especially for important roles. Now, if you look at the key principles for responsible AI in hiring, the first is obviously transparency. We need to clearly communicate on how AI is used in the hiring process to build that trust within the employees. Data privacy. So we need to safeguard candidate information by providing privacy regulations like GDPR and ensuring data protection is happening as well. Also bias. So we need to make sure that the usage of AI is diverse, fair data is used to train the AI, and we are avoiding any perpetual outdated biased hiring practices. So those should be avoided. Also, accountability is there. We need to ensure that the AI decisions can be traced and corrected if errors are being made and someone with someone responsible for oversight. So human intervention is needed in such cases. Also control over AI generated decision, maintaining human oversight over final decisions like candidate selection, even when AI automates most of the tasks like resume screening or interview scheduling. Now if you look at the importance of human oversight in AI driven hiring, there are many like ensuring accuracy. AI sometimes misses on qualifications in a resume or a recruiter can step in and catch it and ensure top candidates aren't overlooked, right, handles complexities. So for unique situations like complex compliance rules, human expertise ensures full accuracy where AI might fall short. So human intervention really helps in such cases. I also building trust. So when hiring managers see that a human has reviewed AI driven shortlist, it boost confidence, especially for critical roles. Also improves the AI. So when human catch errors, their feedback can be given back to the EI to make it more smarter and more reliable over a period of time. Also it enhances quality. For tasks like sending onboarding documents, a human can step in to double check the accuracy, ensuring that nothing gets missed in such cases. Now, if you look at some of the best practices for data privacy in talent acquisition, the first comes as data minimization, which is collecting only the relevant necessary information like skills and work experience, and avoiding any personal details. Then there is purpose limitation, which is using collected data solely for its intended purpose such as screening candidates and avoiding unrelated usage. There is also transparency, which is basically clearly informing the candidates how their data will be used, especially when AI is involved through data through job descriptions or privacy policies. There's also access control, so limiting the access to sensitive data, ensuring only authorized personnel like the HR team can view it. And then the data security, which is primarily protecting candidate data with encryption or password protection is implemented to prevent unauthorized access or breaches. So this is how we can apply these best practices when AI is involved with our HR processes. 35. Introduction and Welcome: Hi, guys. Welcome to this session. In this session, we'll see how we can make use of generative AI and talent acquisition, specifically for smarter candidate screening. In this module, we're going to learn a couple of things, which is going to be first seeing how we can automate the whole screening process and save a lot of time of our human resource. Second is customizing the candidate assessments. So how we are assessing the candidates, we can customize that as well based on their profile. And then how we can streamline the whole workflow. This will really help in reducing or removing redundant processes or steps which we may have in our HR screening process. We will also be trying to answer some critical questions here, like, how can AI driven hiring remain ethical and unbiased? We have to also make sure that that is happening. Second, we're going to see how can you safeguard data privacy and security in AI powered screening process. So we will put in certain steps by which we will get to see how we can do that. And lastly, how do you ensure your team embraces AI as a tool and not as a threat? This has to be the case because that is how you will be able to make use of AI in a productive manner and is generally going to improve the quality of our work. So you're going to see practically different tools like Chat JPT, Gemini, and Claude being used in this module to apply all these concepts. 36. Identify Touchpoints and Opportunities in Onboarding: Hi, guys. Welcome to this session. In this session, we wanted to understand and identify the touchpoints and opportunities in onboarding where we can implement AI. So if you look at the traditional candidate screening process. So the first step is resume screening, where the recruiters go through multiple thousands of applications to assess qualifications. Then comes the initial screening where we verify the basics and we find initial fit for the role or not. And then comes skills and aptitude tests, which can be role specific. So role specific tests are conducted. And then comes the interview process where we provide it provides deeper understanding of the candidate and his or her strengths and weaknesses. Then we do the background checks and references, which helps to verify credibility of the candidate and lastly, there is a candidate selection. So this is the traditional candidate screening process which every company mostly follows. Now, the drawback or inefficiencies in traditional screening is slow process. So as you can see, this is all manual, which takes a lot of time to be conducted and executed by the HRT. Now, this happens also because of the huge volume of recruitment or HR interviews which are being conducted. Thousands of applications pour in whenever there are job openings in a company. And eventually, what leads to this is a delay in hiring. The hiring process becomes much more longer and it takes a lot of time to have candidate selections to be done. Now the other drawbacks or inefficiencies is subjectivity and bias. Because this is humanly done, there can be a lot of uh in the candidate selection, there can be a bias or subjectivity or uh things which the recruiter has not been able to assess properly, which can be missed as a human error. And then also, there is a limited personalization of the candidate in the sense that we try to identify what is the actual strengths of the candidate and trying to find the best fit role for them. So these are all the misses, you can say, in traditional screening. So this is where AI comes into picture wherein generative AI can try to automate the whole candidate screening process, first, wherein we can do it for resume parsing and shortlisting can be done with tools like Chat GPT, Gemini, you can do that. You can also do AI powered screening agents are available Cloud and hATGPT which can help in screening the particular candidate profiles. Then you can also create skills and aptitude tests on these tools like Chat GPT code interpreter, which can be given to the candidates to take, and then we evaluate them on that. Then there is bias detection and fair screening. So here we can make use of Chat JBT primarily to set up a completely bias detecting fair screening process, which does not take any human angle into consideration. And lastly, we can also do personalized candidate feedback based on by uploading the care candidate's resume and giving specific information through tools like Gemini and Chat GPT, where we can give much more personalized feedback to the candidate. So this is how we can use the AI tools to bridge these gaps which we regularly find in our complete screening process in our HR works. 37. Personalized Candidate Screening with Gen AI: Hi, guys. Welcome to the sessions. In this session, we'll talk about how generative AI can be useful to have a personalized candidate screening process, which we can set up with it. Now, when you look at personalizing the onboarding process with CHANGPT, there are multiple things it can help us with. First, it automates the repetitive task, which is going to be screening process. All that can be removed completely and CHAGPT can take over and do that task for us. It will also go ahead and enhance the personalization. Based on the profile, we can personalize the questions and then ask in a much better manner, effective manner. Then that is why it will also improve the efficiency of the whole screening process. We will be able to shortlist the right candidate without any misses. And then also because it is an AI tool, it can also help in reducing bias, which people might have when they are doing this whole process manually. Now, if you look at with respect to how Tangibty enhances candidate screening, so it can send out uh, invitations which are personalized. So it can generate personalized interview invites with the key details the screening candidates, it can create structured role specific screening questions which are much more customized to the profile. It can summarize the interview feedback based on the recruiter input without any bias. It can provide and craft polite and constructive rejection messages as well, which can be really dicey and when humanly done, it can go either way. So it can do in a right manner. And then following up as well, sending out reminders or additional role information can be shared through automation with the help of chatlPT wood. Let's see this in practice. How exactly are we going to do this? Let's say, this is how the interface looks like. What we're going to do here is we're going to provide the tool with JD and a particular CV and ask it to screen the particular CV for the job rule. This is the prompt which we're going to give where we are going to ask you are a recruitment assistant, based on the following GD job description, analyze this candidate's resume and provide a suitable score out of ten. Along with key strengths and weaknesses, the job description and the resume are attached critically evaluate and mention the fitments and gaps. So let's attach We've done that, and now we can provide this. So what the tool is going to do is it is going to look at both the documents and based on which it is going to give us the output. Let's do that once more. So now it is going to look at both the documents, as you can see, it's going to give us an overall score based on that. So it's going to give us the strong fit areas first. Okay? What are the things which are suitable for the role based on the profile given. So digital marketing, digital performance marketing expertise, Google ecosystem, and domain knowledge. So on these, it is giving. And then the key weaknesses as well, partial fit, ownership of the end to and marketing strategy, there is a gap there Okay, so emphasizes on resume emphasizes on training, whereas what they're looking for is end to end marketing strategy, product marketing, lifestyle marketing, noticeable gap, there isn't much experience of that. Okay. So like this, it is going to give on all the aspects, areas which are excellent, exceptional, okay, strong. Okay. And then those which are partial and based on which it's going to give the overall score, which was 8.3 out of ten. So now we have the score here, okay? So let's say additionally what we want to also wanted to do is to create five behavioral interview questions for the given job role that ideally we should ask the candidate that assesses on problem solving, leadership and communication. Okay? And we are also asking Chatb to provide us the ideal answer so that we can compare, right? So if this is already with us, before going for the interview, if we prepare like this, we have the question and the ideal answer as well. And now we just need to compare it with what the candidate says in the interview. So we have the ideal answer as well, what we are looking for to assess on problem solving, okay, specifically leadership, all those particular topics. Lastly, let's say we can also ask AGBT to provide five additional questions on skill based. So specifically for this role, technical questions, basically, which we want to also get. And that also Chat JBT can help us with wherein we can get specific questions relative to topics like formance channel mastery, measurement and KPIs, experimentation and testing, budget management, and so on and so forth. So you can see now with this process, with this approach, you are able to assess the candidate in a much more effective manner and get a concrete idea whether you should be moving ahead with the candidate to the next round or not. 38. Prompt Strategies and Various Gen AI Tools: Hi, guys. Welcome to this sessions. In this session, we'll look at some prompt strategies and various generative AI tools which we can use to apply these strategies. So when you look at the prompt structure, so this is how we need to go about building the structure of our prompts. So there are three parts to it. So the first part is going to be instructions. Instruction is, this is where you tell the AI what needs to be done. So that has to be very clear so that AI knows exactly what it needs to perform and provide the solution. Second is the context. So this defines what the company is looking for, what you are looking for, what information, so you're giving the context, the background of it, what is the background and based on what you are asking the AI tool to do. Then comes the question and output format, which ensures structured responses which we can get from the AI tool in actionable factors, which we can possibly we can produce out of it. These three should be a part of our prompt whenever we are giving them to any AITol. Now there are various prompt strategies which you can keep in mind while writing your prompts. First is pros and cons strategy. This approach helps AI analyze both sides of a hiring decision screening method or evaluation process. By asking the AI to weigh advantages and disadvantages, recruiters can gain a more balanced perspective. The other one is going to be role strategy where you assign a persona to the AI to make its response more targeted instead of giving generic advice. AI responds as if it is in a specific role, such as a senior recruiter at a tech company. Similarly, there can be a Q&A strategy. It helps structure AI's responses by breaking down a prompt into specific questions. This is useful for identifying red flags in resumes, crafting, pre screening questions or evaluating key candidate profiles. Now, another one strategy which you can use is chain of thought strategy. This method guides AI through a step by step reasoning and helping to down the decisions into smaller steps, breaking down decisions into smaller steps, which you can do. These are all different types of strategies which we can really apply. Let's see this how we are going to do this across various AI tools. So what we're going to look out for today is Google Gemini. On Google Gemini, as you see, this is how the interface looks like, you can go ahead and you'll find the settings over here. If you go to settings, you can go to Connected Apps. Connected apps shows you what all other Google Apps products you can link your Google Gemini to. Here you can switch them. You can also switch on additional apps which you want to connect to. Now, let me show you an example of how Google uh Gemini is going to look like. This is where on the top right corner, when you say Ask Gemini, is going to come up in this particular manner where you can go ahead and give your prompt. This is where you're going to give your prompt and you can insert the details in the document. Now let's have a look at an example of how this is going to work out for us. We're going to do a new chat So let's do this particular prompt where we are asking Google Gemini to act as a senior recruiter at a fast growing digital marketing startup. You screening candidates or digital marketing manager roles and which we will attach over here. That requires expertise in performance marketing, SEO, SEM. One candidate has been strong technical skills, but minimum experience working in cross functional teams. How would you assess their suitability for them? Okay? So we're going to attach the documents here the first job description. And then the resume. And now we will ask it to look at the job description, look at the resume based on which, assess the particular job profile and give us a ranking whether the resume is good enough for that or not. So now you can see it has started creating effective summary key strengths, which it has mentioned, potential risks. It is showing evaluation against requirements. It has mapped that also recommendations, technical interview, bypass basic PPC questions, and focus on attribution modeling full for in strategy. So questions which needs to be asked over here, behavioral interview, all these can be asked away. So this is how we can make use of Google Gemini specifically by giving us structured prompt, and it is going to give us the output in the same manner. Let's look at how Cloud is going to work in the same manner. This is how Cloud looks like where we can give the prompt. Okay? So let's give a different prompt now. This is the prompt which we are giving to Claude. We're saying, I need to shortlist candidates for senior data analyst position. The process involves reviewing resumes, conducting rescreening interviews, and assessing technical skills. Guide me through a step by step approach to selecting the best candidates. Break down the criteria, screening methods, and decision making process in a structured way. Now it is going to give us the particular process. First phase is resume review. What is the requirement, experience indicators, create a scoring rubric, Phase two, phone and video screening, which can happen red flags to watch for phase three technical skills assessment, take home assignment, live technical interview which we can call them for and then the final selection where team fit interview can happen, leadership stakeholder interview can happen, final evaluation, and then making the decision. So this is how we can use prompts specifically in a structured manner to get the right output. Another specific thing about these AI tools is that now, if I give this particular prom to Claude, where I say that can you put this in a Crips manner in a tbar format? What this specifically refers to the previous conversation. Okay. So how these AI tools operate is every conversation is going to get stored up in the AIS memory data. Okay. And because of which, whatever questions you may ask, it will refer to the previous conversations and based on which it will respond. So it's like a human conversation experience which you're getting out of it. When you start a new chat, which you have an option of creating on the top left corner on all the AI tools, that will be a new conversation, and the memory will not be referring to that. So every chat will have a specific memory which it will record and based on which the outputs will be given. I hope this makes sense. I hope you're able to understand now how prompts needs to be structured and used on various AI tools. 39. Building a Custom GPT for Resume Evaluation: Hi, guys. Welcome to this session. In this session, we'll see how we can build a custom GPT specifically for resume evaluation. Now, this can be of great use because a custom GPT would be able to analyze the resumes, compare candidate qualifications, scoring candidates. All of this can be done automated with the help of the custom GPT. It will be able to automate the whole resume evaluation process. It will be able to match candidates to job descriptions, which we have provided it also standardizing the whole screening process, and finally, also reducing the hiring bias. So this will really help in, um improving the quality of hiring process which we usually have in our HR departments. So let's see this in practical how we are going to bring this custom GPT. So once you're on on your hat GPT, as you know, custom GPTs are going to be paid feature, so you need to be on a paid version of Chat GPT to access it. So we can go to explore GPTs and we are going to create a new GPT over here. So we're going to give it a prompt. So this is the prompt we are going to give. Create a custom GPT for automating resume screening in high volume hiring, um, especially for software engineering roads. The GPT should include extract key details from resumes, including skills, experience and education, match candidates against job descriptions, highlighting alignment with role specific criteria, score candidates based on technical expertise, experience level, and problem solving skills, ensure bias free evaluation using structured skill based short listing. Generate recruiter friendly reports with suitable scores, key strengths, and gaps. Recruiter should be able to input a resume, receive ranked candidate recommendations and access concise shortlisting reports to sign, speed up the decision making. The knowledge base on this GPT contains the job description. So we'll have to provide the job description as well at the back end of this custom GPT so that it can score the CVs, the resumes based on that. Okay. So this is what we are going to give the custom GPT to create. So let's see how it is going to work. He's going to give it a name as tech resume screener. I'm saying that's fine. It will then generate a profile picture for it for this custom GPT. Once it creates that, then we'll do a testing of it, with actual we'll upload the JD, and then we are going to give the resume. So this is created. Okay, so let's upload the JD as well over here in the knowledge section, job description. Okay. So now we can create this Let's say this is for anyone and we can save this. Now we will go ahead and upload our CV or resume and we ask it to evaluate the resume based on the job description provided at the back end. Let's view this particular custom GPT. This is our custom GPT. So let's upload the CV. And we are saying, evaluate the rest. Is going to look at the JD and based on the profile provided the candidate snapshot given over here, Role fit skills to requirement mapping being done, minimum and preferred qualification. Strengths given over here of the resume, gaps, product marketing, ownership, lifestyle, life cycle or funnel marketing. These are some of the gaps in the resume. Okay, suitability scoring given. So total score provided 90 out of. So now, based on this, as you can see how fast we are able to evaluate the profile against our JD and get specific information based on which we can take our decisions and move forward, uh, with the next steps. This is how we can build a custom GPT for resume evaluation as well, um, and fast our HR processes going forward. 40. Detecting Bias in Candidate Evaluation with Claude: Hi, guys. Welcome to this sessions. In this session, we'll see how we can make use of Cloud primarily to detect bias in candidate evaluation. Now, this can be a very common thing which can happen because of a human error, possibly, where there is a biasness which we find in candidate evaluation. So this is where Cloud can be really useful. It can help to process interview data in a much more structured, concrete manner and detect bias across demographics, provide actionable insights, and standardize the whole process and make it much more the quality of productivity, the output can be made much better with the help of this tool. Let's have a look at it how we can do this. So we're going to use two different datasets, ideally speaking to do the comparison and understand. The first one is where we're going to use this particular prompt where we say, analyze the following interview question, evaluation dataset to detect potential bias in candidate scoring, identify any discrepancies in average scores across different demographics, gender and ethnicity, university, highlight any interviewers whose scoring patterns show significant deviations from the average. Provide a summary of findings and suggestions for ensuring fairer evaluation. Let's have a look at this. So once it does the evaluates the document and let's look at the dataset as well, which we can see over here, ideally seeing, let's have a look at that also. This is the dataset which we are going to use wherein, this is a random dataset which we have created ideally, which we are using in this particular case. So now what happens is it evaluates the whole thing and gives us specific information. Like, for example, the male candidates scored 1.1 points higher on average than female candidates. Okay? So male average, overall score was 6.6, female average score was 5.5, okay? Then it also gives the gender score breakdown. So in technical score, male scored 7.1, communication score 6.1, overall score was 6.6. So the same breakup which it's showing right now. Now, what it gives the analysis is that female candidates, 40% receives lowest score, 4.5, only 10% receives 7.5 plus. Whereas male candidates, 0% received lowest score, 30% received 77.5 plus. No female candidates cute above 7.5. One male candidate scored 8.5. Now it's giving us the analysis in this particular manner, and then it does the ethnicity analysis also university analysis as well and gives us the output. This is a fair evaluation which is being done where we are not able to see any red flags as such in terms of biasness overall, it is going to show us this kind of scoring. Now let's change the data a little bit to understand really the gaps. So the same dataset, we have made some changes, and now we are putting it again into Cloud to analyze. So now this is the dataset which we have used. Okay. So now let's have a look at this, how it is going to work for us. You can see, first of all, systematic gender bias across all interviewers, which we can see over here. So here, what we can see is the I one, specifically, which is first interviewer, male average 7.1 female average is seven gender gap, not much, right? In case of I two also 7.5, 6.5 moderate I three. Now we see it's a severe one which we get to see here. Okay? Extreme scoring inconsistencies. So can two, specifically, we see that and three, there is a lot of gap. Issue three point gap which we can see here, same interview, same gender, same ethnicity, identical performance, three point difference, which we get to see. Bias evidence. We can see bias evidence also cases where candidates with identical performance scores receive different overall scores. And interviewer analysis. Interview analysis, overall, what we see is interview two and Interview three has a poor consistency rating. This is how we can go ahead and identify any kind of biases in our candidate evaluation with the help of EI tools just like Clot. 41. Addressing GenAI Pitfalls in Screening with Human-in-the-Loop Strategies: Hi, guys. Welcome to this session. In this session, you want to discuss about the generative AI pitfalls which can happen in the screening process human in the loop strategies. So what we are trying to say over here is there can be a lot of limitation challenges as well when we are using AI with respect to resume screening process and the HR work. One of them can be requiring human oversight for fairness. So there can be issues with this as well, wherein the AI tool has not been trained properly, and it's giving us the output which has these loopholes. It can also go ahead and generate hallucinations and errors as well unless and until we control it and give proper right instructions, favoring structured resumes unfairly. This can also be a possibility wherein the AI might favor structured resumes not in the proper manner. And overlooking the valid career gaps. If the instructions, the proms are not given properly, then these things can happen wherein it might overlook certain career gaps producing inconsistent assessments. So assessments which are being created or aptitude tests which are created for the profiles are not relevant or not customized to the job skills required. Reinforcing bias in hiring. So it might be a case that we need to look at hiring without any bias as well and lacking explainability in decisions. The decisions which are being given by the AI tool does not have proper explanation or complete explanation. Now, this leads to creating a lot of ethical and legal risks for the business as well if these are not controlled in the right manner. So what we want to do in such cases, we can make use of some reflective prompts which we can give to the output provided by the AI tool and assess it again and check whether the response given by the AI tool are facing these challenges are biased or unbiased. So we want to check that with the help of these reflective prompts. So let's see a practical example of what we are trying to achieve out here. So let's say this is a situation, okay? A recruiter is reviewing an AI generated candidate ranking for a software engineering role and notices that a candidate with strong theoretical knowledge, but no hands on proper experience, project experience is ranked higher than candidates with practical experience, right? So this is not correct. Just because of theoretical experience more, it is ranking them higher, giving them higher points, which should not be the case. So that we want to evaluate. Okay, so let's look at this, and let's also see the dataset which we are discussing out here, ideally speaking. So we are going to Look at a specific data set. Let's say this is the dataset. These are three. And here you can see the first candidate has been given a higher score, which says excellent theoretical understanding of core concepts, strong academic background performs very well in written and oral explanations, but lacks real world hands on project experience, right? So that is the issue which it is facing. Now we want to see whether the AA tool is able to detect this gap specifically. So what we are going to do is we're going to make use of this and we're going to give it a prompt, and we're going to upload the dataset first. And we're going to give the prompt. Does this evaluation align with the actual job requirements that emphasize project experience, or is it influenced by a big list of skills involved in the candidates profile? We want to check whether the AI tool can identify this gap. Okay, alignment with job requirements, the job description proprieties hands on project, end to end delivery, real world problem solving. Then the version should heavily reference specific projects executed, right? Okay? Common red flag in such evaluations, higher scores or positive remarks without clear project evidence, right? What a better bottom line, the evaluation does not fully align with the jobs project centric requirements if it primarily rewards a long list skill list. Right? So now it clearly understands. So this is how we need to go ahead and also a very important thing that when we are using AI so much in our HR works specifically, making sure the output which we are getting is also in the right manner. We should not be looking at AI tool to be used blindly, the output to be used blindly, but we need to do an oversight. Human oversight should always be there, and the AI tool should be used as an assistant in getting the output for us and making our output much more better. The quality of work can be better, but it should not be the one which I'm relying upon. It should not be a case that the AI tools output is what we are dependent upon for the output so that we can use in our process. So, the objective is that we are going to make sure that the prompts to reduce all of this, the prompts need to be much more specific and much more aligned properly with the expectation, which we are giving to the AI tool and get the right outputs. 42. Best Practices and Emerging Tools for GenAI in Screening: Hi, guys. Welcome to this sessions. In this session, we'll talk about the best practices and emerging tools which are happening in GN AI with respect to screening resume and specifically with HR words. If you look at some of the best practices is going to be we have to make sure that ethical AI usage is happening, so the bias monitoring is happening, transparency and explainability increases with AI, and human oversight should remain all the time. Optimizing the AI workflows also for efficiency is going to be there. So where we need to keep fine tuning the prompts which we are using on the tools, automate the data integration should happen so that there is no discrepancy there. And also we are making use of a lot of custom AI agents, which will give us customized outputs. Also, the tools which we are using right now for hiring purposes in AI are going to be cloud and chat GPT works out really well in terms of providing the processes, automating a lot of things through custom GPTs and Gemini and gem as well. Apart from this, we can also make use of Power Automate plus AI Builder to automate these processes and build tools which can generate high quality output. Now, other than this AI techniques which are transforming hiring right now, if you look at it, there are AI generated behavioral insights. So a lot of the information which we are getting, so understanding how it is the behavioral insights which we're getting from AI can be useful as well. There is also bias detection algorithm. So now we have custom GPT, which can detect bias unbiased outputs. So that is also coming up in the future. So there is predictive hiring models. A lot of hiring models are being created based on AI, which can be predictive in nature, giving us much more leverage about hiring high quality candidates for our businesses. Then there is also AI analysis on video interviews, specifically giving us inputs from there and making us able to understand the candidates strengths and weaknesses. Now, in order to because this is going to be ever growing and new tools are coming in, we need to stay ahead with this AI usage in HR specifically, wherein we need to keep learning about these tools and use them on a regular basis, improve our prompt engineering as well so that the quality of prompts which we're giving to these tools are also very precise and accurate so that we get the desired results. Need to keep experimenting and iterating with different types of proms, custom GPTs which we can create, which will really help in giving us much better HR outputs, and also making sure while we do these, we adopt ethical AI practices all the time so that it gives us the right output, and we are using the AI technology in the right manner. 43. Introduction to Legal Considerations: Hi. Welcome to this session. In this session, we wanted to understand the legal considerations which we need to have in AI with respect to HR. So AI and HR has a lot of initiative right now, and if you see AI is reshaping HR for recruitment to performance management. In all the areas, AI can be integrated. But with innovation comes a lot of complexity. So issues like data privacy, bias prevention, and accountability need careful intervention and legal navigation. So the compliance of with regulations like the GDPR, General Data Protection Regulation, the California Consumer Privacy Act, CCPA, and anti discrimination laws are becoming critical. Now, discrimination and bias in AI has the power to transform hiring, right? Now, but it can also perpetuate biases if unchecked. So that's why discriminatory algorithms are a legal risk. Now, to prevent this, organizations must audit their AI systems regularly ensuring fairness and transparency in all HR processes. There's also, if you see, privacy is becoming paramount. AI systems handle vast amount of sensitive employee data. To comply with GDPR and CCPA, all organizations must secure explicit consent, maintain the transparency and protect personal data. Only relevant data should be collected, no more, no less. Now, with respect to this, there are some ethical and legal practices as well, which we should be applying. So real world examples show that when ethical and legal practices are embedded from the start, the AI and HR can be really transformative. Successful organizations have embraced EI audits, legal risk assessments, and clear documentation with which they're to do minimizing the risk and building a lot of trust. As AI continues to evolve in HR, legal considerations remain at the forefront. Companies must be proactive in addressing privacy, discrimination, transparency, and consent. With the right ethical frameworks in place, AI can revolutionize HR in a legally compliant, socially responsible way. By integrating legal and ethical considerations into AI practices, we can ensure a fairer, more transparent and efficient future in HR. So I hope this is how you understand how legal considerations are going to be in the HR policies. I 44. Data Protection and Privacy Laws: Hi, guys. Welcome to this session. In this session, we wanted to talk about the data protection and privacy laws. So in the key regulations like the GDPR and the California Consumer Privacy Act CCPA, set the standards for transparency, consent, and data security. Understanding these laws helps helps organizations protect individual privacy and maintain compliance in a complex digital environment. So let's look at how GDPR works. GDPR or General Data Protection Regulation is a comprehensive law, effective EU EA and global organizations focusing on lawful, fair and transparent data processing. GDPR principles include lawfulness, pursue limitation, data minimization, accuracy, storage limitation, transparency, and confidentiality, ensuring responsible data handling by organizations. These principles also grant individual significant rights over their personal data and under GDPR. GDPR grants rights like access, rectification, erasure, and data portability, empowering individuals to control how their personal data is used. Now, there are some compliance steps as well. Organizations achieve GDPR compliance through data protection impact assessments, DPIAs basically appointing a data protection officer, implementing privacy by design, and obtaining informed consent before processing any data. Similarly, let's look at how CCPA works. CCPA, which is the California Consumer Privacy Act, empowers Californians with rights like data access, deletion, opt out options, and protection against discrimination. To ensure these rights, the businesses need to follow specific compliance measures under CCPA. Businesses must provide clear notices, handle data access, and deletion requests promptly, train employees and update privacy policies to comply with CCPA. Beyond GDPR and CCPA, various global data protection laws also play a crucial role in safeguarding personal data. Other significant laws include personal Information Protection, Electronic Documents Act, PIPEDA in Canada, Protection Data Protection Act in Singapore, Australian Privacy Act, and Brazil's General Data Protection Law. Each with significant requirements, unique requirements for protecting personal data. Understanding and complying with global data protection laws like GDPR and CCPA is essential. It's important for safeguarding personal data. By adhering to these regulations, organizations can protect privacy and build trust in a rapidly evolving digital landscape. 45. Employment Law Implications: Hi, guys. Welcome to this session. In this session, let's look at employment law implications due to AI, right? So implications of AI in employment can be multiple. So AI streamlines HR processes, but also possesses a lot of legal risk. Key challenges include potential discrimination, data privacy concerns and the need for transparency in AI driven decisions. One of the most pressing concerns in AI driven HR is the risk of discrimination and bias. AI systems may unintentionally perpetuate biases from historical data leading to discriminatory outcomes. HR must ensure fairness in AI driven resume screening and performance evaluations. Also, alongside bias, data privacy and security are critical areas that demand attention. AI's reliance on vast employee data sometimes raises privacy issues. Organizations must comply with regulations like GDPR to safeguard personal data against unauthorized access and breaches. Beyond privacy, transparency in AI processes is essential for trust. Employees deserve to know how AI impacts their careers. Organizations should make AI processes transparent providing understandable explanations for AI driven decisions. Ethical considerations also play a significant role in the responsible use of AI. AI and HR, such as employee monitoring raises ethical questions. Employers must balance technological benefits with respecting employees rights and privacy. Now if you look at the legal framework for AI and employment, there is anti discrimination laws and privacy regulations guide AI use in HR. Organizations need to ensure compliance incorporating accountability and transparency in AI practices. There's also risk mitigation, which is HR can mitigate legal risk by adopting robust compliance measures, ethical guidelines, and ongoing collaboration with legal experts. Training programs can enhance AI literacy among HR professionals, ensuring responsible AI usage. So at the end, AI brings a lot of efficiency to HR but also introduces serious legal and ethical challenges. Addressing risks like discrimination, data privacy and transparency is essential for responsible usage of AI. Organizations must comply with employment laws, adopt strong compliance measures, and collaborate closely with legal experts. Ongoing trainings ensure HR professionals can manage AI tools ethically and lawfully building trust and fairness in the workplace. 46. Conducting AI Audits: Hi, guys. Welcome to the sessions. In this session, we'll see how we are conducting AI audits. So EI audits systematically evaluate EI processes to identify and address potential biases, ensure data privacy compliance, and build trust through transparency. Effective compliance measures help organizations mitigate legal and reputational risk while leveraging AI responsibly in HR. Now, AI audits systematically evaluate whether AI systems meet legal and ethical requirements. In HR, these audits ensure that AI driven processes are fair, transparent, and accountable. Key aspect of AI audits is ensuring fairness across all HR activities. And AI audit play a vital role in eliminating bias in HR processes. By reviewing algorithms, they prevent discrimination based on attributes like race, gender, or nationality. Beyond fairness, compliance with laws is also a critical component of AI audits. Organizations must comply with laws like GDPR when using AI and HR. EI audits verify adherence to these regulations, ensuring data protection and privacy are maintained. Now if you look at the transparency and trust in AI systems, transparency is another essential factor in building trust in AI systems. AI audits assess the transparency of decision making processes. Transparent systems help employees understand how employment related decisions are made, fostering trust in AI. Now, mitigating risk is another reason why AI audits are indispensable. Neglecting AI audits can lead to lawsuits and damage to reputation. By identifying and resolving issues early audits protect organizations from legal and reputational harm. Now let's look at how the AI audit process happens. The I audit process involves identifying AI systems, establishing compliance metrics, collecting and analyzing data, evaluating decision making, and implementing corrective measures as needed. Monitoring is key to sustaining the compliance over time. AI audits are important for maintaining fairness, transparency, and legal compliance in HR. Continuous monitoring and documentation are necessary to adapt to evolving ethical standards. This ongoing vigilance ensures EI continues to support ethical and effective HR practices. To ensure legal compliance and fairness in AI driven HR, organizations should conduct regular EI audits. These audits assess fairness, transparency, and adherence to privacy laws like GDPR, helping mitigate risks and protect against legal and reputational damage. Continuous monitoring is essential for sustained compliance. I hope this makes sense. I was able to understand how AI audits are so critical in making sure that the policies are in place and the AI usage is done in much more ethical manner. Um, 47. Risk Assessment and Mitigation: Hi, guys. Welcome to the sessions. In this session, we'll talk about the Risk Assessment and mitigation with respect to AI and HR. Now, if you look at Risk Assessment, helps organizations identify and evaluate potential legal risks associated with AI and HR, such as bias, discrimination, data privacy, breaches and intellectual property concerns. Now effective mitigation strategies involve implementing safeguards, updating policies, and collaborating with legal experts to minimize these risks. Continuous monitoring and review are essential to maintain compliance and adapt to evolving regulations. Now the importance of risk assessment is that conducting these risk assessment allows organizations to anticipate legal issues like buyers discrimination and data privacy breaches, taking proactive steps to safeguard employees and customers. Now, steps involved in risk assessment are the process starts by identifying potential legal risks such as security breaches or intellectual property infringement, followed by evaluating their likelihood and impact on the organization. Next, we look at how to evaluate the AI systems and potential AI systems we are using. Now, analyze AI systems to understand algorithms, data sources and potential bias, identifying areas where legal risk may emerge, ensuring compliance with regulations. Mitigating these legal risk, once risks are identified, it's important to develop strategies to manage them effectively. Develop risk mitigation strategies based on identified risks, implement safeguards, update policies, and partner with legal experts to manage and reduce potential legal threats. Now, with respect to monitoring review, continuous monitoring and review the effectiveness of mitigation strategies needs to be done. Stay updated on legal changes to ensure ongoing compliance and adjust strategies as necessary. Now, if you look at the key legal risk in HR AI is common legal risk includes buyers data privacy issues, employment law impacts, and intellectual property concerns. Addressing these risk is vital for successful AI implementation in HR. Now, there are some proactive compliance also proactively managing these legal risks through comprehensive risk assessment ensures that AI and HR operates within legal boundaries, protecting both the organization and its stakeholders. So finally, if you look at it, effective risk assessment in AI for HR involves identifying potential legal risks like bias, data privacy breaches, and intellectual property issues. By evaluating these AI systems, developing mitigation strategies, and monitoring compliance, organizations ensure that AI operates within legal boundaries and safeguarding the stakeholders. I hope this makes sense. I hope you able to understand now what are the different types of legal risks you may face with respect to AI in HR. 48. Documentation and Transparency: Hi, guys. Welcome to the sessions. In this session we'll talk about documentation and transparency. From data sources to decision making clear records, build trust, fairness, and accountability. Detailed documentation of data collection methods, AI models, training processes, and decision criteria ensures transparency and enables effective audits. Now, there is the importance of data documentation. If you have detailed documentation of data sources, collection methods, and cleaning procedures is vital. It helps identify potential biases, ensuring fairness and transparency in AI driven HR decisions. If you look at the model documentation, transparent model documentation includes details on algorithms, hyper parameters, and performance metrics. This clarity allows understanding, scrutiny, and bias detection in AI models used in HR. Also, when we look at training and validation documentation, the training and validation processes are equally important. Documenting training and validation processes includes datasets and techniques, ensures transparency. It enables replication and verification, ensuring the reliability of the EI models in HR. Also the decision making documentation, which is where reporting decision criteria, factors considered and threshold set ensures transparency in AI decision making. This documentation is essential for audits and assessing fairness in HR decisions. Accountability and auditing when we look at that, transparent documentation enables organizational accountability and ensured auditing. A clear trail of information supports ethical AI usage in HR, ensuring adherence to best practices and standards. Also, when we look at mitigating biases in AI, transparency in documenting data and models helps identify and mitigate biases in AI algorithms, promoting fairness and reducing the risk of discrimination in HR practices. Finally, with thorough documentation of data sources, AI models, and decision making processes is crucial for transparency and accountability in HR. Where recording these training methods and bias mitigation strategies, organizations ensure fair, ethical and legally compliant, AI driven HR practices. 49. Keeping Up with Regulatory Changes: Hi, guys. Welcome to the sessions. In this session, we'll talk about how we can keep up with regulatory changes. So as technology advances, so must our understanding of the legal and ethical boundaries it operates within. Staying informed isn't smart enough. It's essential for ensuring fairness, compliance, and organizational safety. So why regulatory updates are important? As EI technology evolves, it is essential to ensure compliance with the latest legal and ethical standards. Regulatory updates define the boundaries within which EI operates, helping HR professionals ensure compliance and avoid costly legal consequences. These updates promote fairness and equality in EI driven HR processes like recruitment and performance evaluations. Now compliance and risk management, these updates also offer valuable insights into potential AIRS. Staying informed helps HR professionals manage those risks safeguarding both the organization and its employees. Now, when you look at strategies for staying informed, first, what you can do is subscribe to newsletters from regulatory bodies. This ensures you get timely updates directly in your inbox. Second, you can follow regulatory agencies on social media to engage in real time discussions and updates on platforms like Linden and Twitter. Third, you can attend webinars and workshops to learn from experts and gain in depth insights into regulatory changes. And lastly, you can join professional associations like SHRM and engage in online communities where professionals share experiences and advice on the latest regulations. Some best practices which you can keep in mind are you can start by establishing a compliance team to monitor and address regulatory updates regularly. You can ensure your EI systems align with the latest regulations by evaluating them regularly. If there are gaps, you can take prompt action to correct them. Developing training programs to keep your HR staff informed about regulatory changes and their impact on EI practices. And finally, consider external audits to identify any compliance and blind spots and benchmark your practices against industry standards. So finally, by staying proactive and informed, HR professionals can effectively navigate the ever changing regulatory landscape. This not only ensures legal compliance, but also fosters trust and fairness in AI driven HR processes. 50. Stakeholder Engagement: Hi, guys. Welcome to the sessions. And this session we'll talk about the stakeholder engagement. Effective collaboration among HR professionals, legal experts, data scientists, and employees helps organizations stay aligned with evolving regulations while fostering transparency and accountability. Engaging stakeholders early and maintaining open communication channels supports ethical EI usage and strengthens trust across the organization. If you look at understanding stakeholders, stakeholders in AI for HR include HR professionals, employees, legal experts, and data scientists. Their diverse perspectives contribute to a comprehensive compliance strategy. Now gaining compliance insights, first, it allows organizations to gain valuable insights into compliance risks and challenges. By involving stakeholders early, organizations can identify potential issues and ensure a comprehensive understanding of compliance needs. Secondly, it fosters a shared sense of responsibility. When stakeholders are engaged in compliance discussions, they take ownership of the process helping to drive a culture of compliance within the organization. Lastly, it promotes transparency. Open communication across different stakeholders ensures that AI related concerns are addressed, building trust, and facilitating the resolution of compliance issues. Effective stakeholder engagement. First strategy is to identify and involve relevant stakeholders. With stakeholders such as HR specialist, legal teams and data scientists early in the process, their perspectives are critical for addressing compliance concerns. Secondly, establish clear rules and responsibilities, clearly define the roles and roles of each stakeholder. Avoids confusion and ensures that everyone knows how they contribute to compliance goals. Third, foster open dialogue and communication. Encourage open communication through regular meetings and workshops. This creates a space for stakeholders to voice concerns and collaborate on solutions. For, share knowledge and best practices, create platforms for stakeholders to share knowledge. This helps everyone stay informed and improves compliance across the organization. Five, regularly review and assess compliance measures, assess the effectiveness of the compliance measures and adjust as needed. Finally, stakeholder engagement is essential for ensuring that EI and HR is compliant, ethical and transparent. By collaborating effectively, organizations can navigate the regulatory landscape and build a culture of compliance. 51. International Considerations: Hi, guys. Welcome to this sessions. In this session, we'll talk about the international considerations. Navigating diverse legal, cultural, ethical landscapes is essential for organizations operating globally. Key regulations like general data protection regulation, GDPR, CCPA, LGPD and International Labor organizations standards, sets requirements for data privacy, fairness and transparency in AI driven HR processes. Implementing robust compliance assessments, data governance policies, and ongoing training helps organizations manage these complexities effectively. Now when you look at GDPR compliance in EI for HR, as EI and HR grows, organizations face various international regulations governing data privacy, discrimination laws, and ethical EI standards. Now, prominent regulations include the GDPR in Europe, CCPA in California, and LGPD in Brazil. Each of these shapes how personal data is handled in AI driven HR processes. Now, looking at the global regulation, understanding and complying with these global regulations is important for mitigating legal risks, maintaining fairness, and building trust with employees and candidates. Adhering to global regulations is vital. Non compliance can lead to severe legal consequences, reputational damage, and ethical dilemmas. By staying informed about international laws, organizations can foster responsible EIU in HR while avoiding legal pitfalls. Some examples of Glogal regulations, GDPR Europe governs personal data protection, requiring strict consent and security measures for processing employee data in AI systems. CCPA USA protects the privacy of California residents, mandating transparency and control over personal data. LGPD Brazil regulates data usage, ensuring privacy and security for Brazilian employees. ILO standards protect employees' rights and prevent discriminatory EI practices globally. These regulations guide how AI systems should process data and make decisions in HR while promoting fairness and transparency. Some best practices for global AI compliance can be conducting a comprehensive compliance assessment. Regularly review global regulations to ensure alignment with AI systems in HR, collaborate with legal experts for regional insights, implement data governance and privacy policies, develop clear robust data governance practices and privacy policies that comply with laws like GDPR and CCPA to safeguard personal data. Continuous monitoring and training, regularly monitor, compliance, and train HR staff on global regulatory standards, data protection, and ethical AI practices. Ensure transparency in AI decisions and create ethical guidelines to foster trust within your organization. Following these best practices, organizations can navigate the complexities of global regulations and leverage EI responsibly in HR. Navigating international regulations is crucial for responsible EI use in HR. Complying with regulations like GDPR, CCPA, LGPD and ILO standards ensures data privacy, fairness and transparency. By implementing best practices such as regular compliance checks, robust data governance, and ongoing training, organizations can make sure manage legal risks and maintain ethical practices. 52. Aligning Ethical and Legal Considerations: Hi, guys. Welcome to this sessions. In this session, we'll talk about aligning ethical and legal considerations. AI offers powerful advantages, but its use must align with ethical standards and legal requirements. We will examine key principles like fairness, transparency, data privacy, and legal compliance. So transparency and explainability. AI technologies offer numerous benefits in HR from streamlining processes to improving decision making. However, to fully harness AI potential, organizations must align their practices with ethical standards and legal regulations. This alignment ensures fairness, transparency, and protection of employee rights. Ethical considerations in HR AI focus on fairness, transparency, and safeguarding employee welfare. EI systems should be understandable with clear explanation of how decisions are made and what data is used. This foster trust and allows employees to challenge unjust decisions. EI systems must be free of bias. Regular audits and fairness metrics help ensure that AI driven HR decisions do not disproportionately affect any group based on gender, race, age, or disability. Personal data must be handled responsibly. This includes obtaining consent, ensuring security and anonymizing data when possible while complying with global privacy laws. Now, there are some legal alignment as well, employment laws and regulations. AI systems should adhere to laws governing hiring practices, employee rights, and benefits. Non compliance can lead to significant legal risk. Intellectual property rights, AI and HR must respect copyright licensing and patents. Organizations must ensure that proprietary AI technologies are protected and do not infringe on others intellectual property. Data governance and cross border compliance with global operations, organizations must navigate various data protection laws. This includes ensuring that cross border data transfers comply with regulations like GDPR in Europe and CCPA in California. Intellectual property and data governance to achieve ethical and legal alignment, organizations should implement comprehensive compliance assessments to stay aligned with evolving laws and ethical guidelines. Adopt robust data governance frameworks to ensure data privacy and security. Conduct regular audits to identify and mitigate bias in AI systems. Provide ongoing training to HR teams on legal and ethical standards, ensuring consistent application across the organization. Effective AI use in HR depends on aligning technology with both ethical principles and legal requirements. Organizations must prioritize fairness, transparency and accountability to prevent bias and protect employee rights. Finally, clear policies, ongoing training, and collaboration with legal experts help ensure compliance and build a culture of trust. By integrating ethical and legal standards into AI development, organizations can build trust, minimize risk, and create a fair, transparent HR environment. 53. Developing Ethical and Legal Guidelines: I Hi, guys. Welcome to this sessions. And this session, we'll talk about developing ethical and legal guidelines. So with AI reshaping recruitment, performance evaluations, and more, it is essential to ensure fairness, transparency, and compliance. We will explore the principles and laws that help build trustworthy human centered EI systems in HR. So to ensure AI systems in HR are deployed responsibly, it is vital to address both ethical and legal considerations. Ethical concerns include fairness, transparency, privacy, and preventing bias in decisions like recruitment and performance evaluations. Meanwhile, legal considerations revolve around adhering to data protection laws like GDPR, CCPA and anti discrimination regulations. Together, these considerations help build trustworthy, fair and compliant HR systems. Now, when you look at key ethical principles, the first comes fairness. AI algorithms must avoid bias and ensure fairness, essentially concerning protected characteristics like gender, race, or disability. Bias. AI systems are prone to bias and whether from bias data or flawed algorithms. To mitigate this, diverse training data and regular audits are essential. Interpretability, HR professionals must ensure AI systems are transparent, though decision making logic should be clear and understandable to all employees. Privacy. Protecting personal data is critical. AI and HR must adhere to data protection regulations and adopt privacy by design practices. Consent always obtain informed consent from individuals when collecting and processing personal data in AI driven HR processes. Legal considerations in AI for HR, data protection. AI systems must comply with data privacy laws like GDPR and CCPA. Personal data must be securely stored and processed. Anti discrimination. AI tools must not perpetuate bias in hiring or performance evaluations. Adhering to non discrimination laws ensures fairness in HR decisions. Employment laws. AI systems should align with employment regulations to protect workers' rights, particularly around issues like performance monitoring or employee development. Developing ethical and legal AI guidelines. So identify another identify stakeholders involved HR professionals, data scientists, legal experts, and employees to create comprehensive guidelines. Assess ethical and legal risks, identify risks like bias decisions or privacy breaches and address them proactively. Define ethical principles, establish principles such as fairness, transparency and privacy to guide all AI system development. Review legal frameworks. Familiarize yourself with laws like GDPR, CCP, and anti discrimination regulations to ensure compliance. Establish data governance, develop frameworks for responsible data collection, storage, and usage, ensure privacy protection and data minimization. Design accountability measures, implement auditing process and mechanisms for individuals to challenge EI decisions. By establishing strong ethical and legal guidelines, organizations can ensure AI systems in HR are fair, compliant, and transparent these guidelines foster trust, reduce risks, support, responsible AI deployment across all AI, HR functions. With clear accountability and ongoing training, organizations can maintain ethical standards and legal compliance as AI continues to evolve. 54. Case Studies: Hi, guys. Welcome to this sessions. In this session, we'll explore and look at some of the case studies where we have gone ahead and used implemented AIN HR. So we're going to look at three inspiring case studies that demonstrates how organizations successfully align their AI driven HR systems with ethical and legal standards. These examples showcase how challenges in bias transparency and privacy were addressed through careful strategies. So let's look at the first case study. XYZ corporation implemented an AI powered recruitment tool, but found that the system favored certain demographics leading to biased hiring outcomes. The training data which reflected historical hiring patterns reinforced these biases causing discrimination based on gender, race, and educational background. XYZ or Corporation took proactive steps to address the bias. This started with a thorough audit of the historical recruitment data to identify bias sources. They expanded the training data set to include a more diverse range of resumes, ensuring a fairer representation across demographics. Additionally, they incorporated fairness constraints into the AI model and used explainable AIXAI techniques to make the models decisions transparent. They also introduced human AI collaboration where recruiters made the final decision after reviewing the AI's recommendations. Further bias awareness training was training for recruiters helped reduce human induced biases. The AI systems bias score improved by 40%, reflecting a significant reduction in biased outcomes. The diversity in the talent pool increased by 25% within six months, and efficiency improved with a 30% reduction in time to time to hire. This success boosted internal trust and ensured legal compliance with anti discrimination laws. Let's look at another case study. At ABC Inc, the traditional employee evaluation system was perceived as opaque and biased. Employees raised concerns about favoritism and inconsistent evaluation criteria, creating a lack of trust in the performance review process. So ABC Inc tackled these challenges by designing an AI driven evaluation system with a focus on transparency. They used XAI technology techniques to make performance evaluations understandable to employees. A dashboard was created to provide employees with clear insights into their performance scores, contributing factors and areas for improvement. The company also integrated bias mitigation strategies, auditing the AI system regularly to ensure fairness across different employee demographics. ABC piloted the system in one department before rolling it out organization wide, ensuring a smooth transition. As a result, 90% of employees reported a clearer understanding of their evaluations and employee satisfaction with the fairness of performance reviews increased by 35%. The transparency and feedback led to a 20% boost in productivity as employees would focus on specific areas for improvement. The system helped ABCing foster a culture of trust and accountability. Let's look at another case study three, where DEFCOp adopted an AI driven payroll system to streamline payroll processing. However, employees expressed concerns over how their sensitive personal data, including salaries and identifiers, would be handled securely fearing potential breaches. DEFCOp prioritized data privacy by implementing a robust data protection framework. They conducted a privacy risk management assessment and employ data anonymization techniques to protect employee identities. Additionally, they integrated advanced encryption protocols for storing and transmitting sensitive payroll data. They restricted access to this data through role based permissions and multifactor authentication, which is MFA. To enhance transparency, they used blockchain technology to provide immutable audit trails for all employee transactions, allowing employees to track how their data was being used. Federated learning was applied to train the AI model on decentralized data, ensuring privacy while still benefiting from AI analysis. The measures resulted in zero data breaches in the first year and 85% of employees expressed confidence in the system's ability to protect the policy, their privacy. Payroll processing time was reduced by 50%, improving operational efficiency. These efforts not only safeguarded sensitive data, but also enhanced employee trust and compliance with global data protection regulations. So finally, these case studies showcase how organizations can address ethical and legal challenges in AI for HR. By focusing on bias mitigation, transparency, and privacy detection, companies can successfully align AI systems with both ethical standards and legal requirements. 55. Thank You For Taking This Class!: Hi, guys. Congratulations for coming to the end of this class. Thank you for taking this class. I hope the content was valuable, able to understand now how we can use these AI tools to integrate it in our day to day HR works, and I hope you're able to implement them practically in your business and for your clients. Thank you once again for taking this class and I'm really excited to see you again in a new class, so.