For this project, I used NotebookLM to synthesize and manage the literature and technical documentation for my master's dissertation on carbon-optimal AI inference scheduling.

Here is how I applied the tool's features to my workflow:

• Source Grounding: I uploaded several core research papers covering predictive analytics, machine learning, and sustainable grid management to create a centralized knowledge base.

• Data Exploration: I included the documentation and metadata for the PGCB Hourly Generation Dataset. This allowed me to easily query specific constraints and variables while drafting my proposal without having to dig through raw files.

• Rapid Synthesis: By using the chat interface, I was able to generate summaries comparing different approaches to AI energy meters and load management. This helped me identify gaps in the current research and efficiently structure my literature review.

Overall, this setup has significantly streamlined how I interact with dense academic sources and technical datasets.