Exploring LLMs and RAG at the 2025 AI Winter School (Brown University)

I recently gave a 2.5-hour hands-on workshop on Large Language Models (LLMs) at the 2025 AI Winter School, hosted by the Center for the Fundamental Physics of the Universe at Brown University. This session focused on setting up an LLM with both the OpenAI API and an open LLaMA model, as well as building a basic retrieval-augmented generation (RAG) system to demonstrate how researchers can leverage specialized, domain-specific information about their research when using LLMs.

During the workshop, participants worked hands-on with both API-based and locally hosted models. They left with practical insights into how to integrate LLMs and RAG into specialized research and engineering tasks, particularly those involving quantitative or highly technical data.

Workshop Recording

You can watch my full 2.5-hour workshop here:

• Workshop Recording

Jupyter Notebooks

For a deeper look at the code and examples, check out these notebooks:

• OpenAI-based LLM Setup
• LLaMA-based LLM Setup

Slides

You can also review the presentation slides for additional context:

• Slides (PDF)

Event Page

For more details on the 2025 AI Winter School, including the full schedule and other workshop sessions, visit the official event page:

• 2025 AI Winter School Event Page