DeepSeek Mastery: An Introduction

1. Introduction: Hi, welcome to this Deep Seek mastery class. Whether you are a developer, content creator, marketer, business owner, or any other professional. This class is your great way to unlocking the full potential of deep Seek. In this class, we will understand what is deep Seek and what are the powerful features we have in this tool. Plus you will understand what is prompt engineering and prompt engineering techniques in a very simple language. You know, AI is evolving very fast, too, so enroll in this class to learn this new technology. Thanks for checking out this intro. I hope to see you inside the class. 2. What is DeepSeek: Hi, welcome, everyone. So what is deeps? Deep Sk is a chatbot just like Cha JBT, but it seems to be more powerful than Cha JBT because it is solving complex problems of coding and mathematics and providing easy explanation of the solutions it generates, and it is performing really well on these benchmarks. Also, you can see that it is free to use. So you just need to click on Start now. And here you have chat.deeps.com. You can access it here and just start typing your prompts. So this is very simple to use. You just need to type in your prompts here. So let me type in suggest to me some ways to help people with native narrative AI tools. Let's press Enter. Here we have 12 different categories, and each category has some subcategory, here you can see that. So I can use AI to help non profits by generating grant proposals, fundraising emails or social media content. Or I can go for writing support or personalized tutoring. So this is generated by deepsk. If I will have to think about this on my own, then it won't be possible to get these many ideas. So this is the power of AI, and we will understand how we can leverage Deep Seek so that we can grow professionally and personally. Now, thanks for watching this lecture. I'll see you in the next lecture. 3. DeepThink R1 and Search Option: Hi, welcome, everyone. Now in this lecture, we will understand how we can use Deep think ARVN to solve reasoning problems and how we can use this search often. So let me click here first. It means that ARVN model is active now. Now I will copy a problem from here and paste it in deeps. Unpleasant. It is thinking. Now, see here, the power of the the th model. It took one 91 seconds to solve this problem and it went into every detail and every alternative. Here you can see this alternative. Then we have this alternative and this one a lot of thing went in this solution here. It is very detailed. I don't believe that any other AI tool can do this nGVT can't do this, and this is the solution. This part is the deep thinking. We can see that here it is very long and then we have this solution and it is two. The answer is two. Now let's check here. The correct answer is two times. This is the power of deep think R one by deep sik, and this is going to help you make better decision based on the explanation you get here. Now let's see how we can use this search option. It is used just like search engines. So just click here. And since DC do not have access to Internet, so let's type in what is the time in New York City right now. Here you can see that we have the time, but we also got some additional information that provided us time difference with Beijing because Deep CQ is a Chinese company. Now let's search something else. Let's say influential marketing for air tools. Now, here you can see that too. I did not search the web, just like it did here because if you put your cursor here and click, it says, search the web when necessary. So it was not necessary for DeepCk to search websites for influencer marketing for AI tools, which means that Deepsq will only search websites if DeepCq is not capable to help you with your prompt. So this is all about DT ARVN and the search option we have here. If you have any questions, then please message me anytime. Thanks for watching this lecture. I'll see you in the next lecture. 4. Extract Info with DeepSeek: Hi, welcome, everyone. In this lecture, we will understand how you can use deepsk to extract information from BD files and images. So to upload your file, so you will have to click here. And it says text extraction only. And you can upload only dog files and images. And the maximum number of files that you can upload is 50, and each file size should be less than or equal to 100 MB. Now, let me show you the files I'm going to use for this demo. So I'll use this European AIA file and the image that we have on the website of Deep Seek. So let's go to Deep Seek and click here. Now let's click on this image and upload it. Now I will ask you DCq to explain this image to me. So just type in explain this image to me and press enter. So this is the explanation, and I know that it was a comparison between Deep seek model and Openes model. So this is explanation of the graph in this image. Now let's upload the PDF file. So here it is. Now, let me go back to the file again. Here you can see that we have some points one, two, three, about European AIC. So I'll ask Deep Seek to create a table. So let me type in create a table for the information we have in this PDF file. Presenter. It is throwing an error saying the server is busy. You just need to click on regenerate, and it might work this time. Again, the server is busy, so I will click on regenerate again. This is the problem we used to face with Chat GPT. When it was launched and the same thing is with deepsk. Now on the third attempt it generated the output for me. So this table organizes the information into clear sections for easy reference. Now you can easily understand the PDF that I uploaded here. So it says all AI systems deployed in the European area, including those developed or bought from third parties, and rest of the points are also explained very clearly here. The same way, you can summarize any other type of document that you may have, like research papers or proposals, or balance sheets. If you have any questions, then please message me anytime. So that's all about text extraction from images and dog files. Thanks for watching this lecture. I'll see you in the next lecture. 5. What is Prompt Engineering: Hi, welcome, everyone. In this lecture, we will understand what is prompt engineering with a very simple definition. So prompt engineering is the process of designing and crafting the quotients or statements given to Deep Seek as input in order to get specific and desired outputs. It involves understanding the capabilities and limitations of the deep Seek model and using that knowledge to craft prompts that will generate high quality and accurate responses in your favor. I've written this definition personally so that everyone can understand prompt engineering in just a minute. And I hope that you understood it well. But still, if you have any questions, then please message me anytime. Thanks for watching this lecture. I'll see you in the next lecture. 6. Prompt Engineering Techniques & Project: Hi, welcome everyone. In this lecture, we will understand prompting techniques. So the first one is zero shot prompting. In this type of prompting, we simply input a prompt in a form of quotien or statement, and we do not provide any context behind our quotien or statement. The second technique is role prompting. In this technique, we provide a role to the chatbard and then we ask a question or we can use a statement like this. So here is a rule and we have this statement. Now the next technique is few short prompting. Here we create a setup like this. So in case of ing, we provide a quotien then it's answer, then a quotien then it's answer. And then we ask the question for which we want the chatbard to create an answer. And for statements, we use the same format, but in the form of statements. And here you can see the last statement is also complete. So basically, this setup is the context we are providing to chatbot models. Now, let's go to the last prompting technique. Which is chain of thoughts prompting. So in this technique, if we are using a quotien, then we provide the answer to that quotien as a setup or the context. And in case of statements, we provide the context, but we don't completely finish the last sentence. So the chatbot will have to think in terms of these chain of thoughts and generate the response accordingly. I hope that now you understand all the prompting techniques. Now I also want to give you a project. So what you have to do is you have to create a prompt for each prompting technique from your specific industry. And trust me, when you will work on this project, this will really help you to create better prompts. So thanks for watching this lecture. I'll see you in the next lecture. 7. Skillshare Thank You: Hi, thanks for completing the class. I hope that you enjoyed the content. And if you have any questions related to deepsk or prompting techniques or the prompts I shared in the prompts document list, then please ask your questions anytime. Again, thanks for watching this class. I wish you success in your professional and personal life.