This hands-on workshop introduces researchers and students to building their own AI-powered research assistant using retrieval-augmented generation (RAG) — a method that combines large language models (LLMs) with your own datasets. Participants will learn how to connect and query diverse sources of health, environmental, or behavioral data to accelerate discovery, evidence synthesis, and decision-making. Using GPU-accelerated tools and guided examples, attendees will explore Python code and experiment with building intelligent assistants that can summarize literature, extract findings from reports, and support research workflows in population, clinical, and community health. Registering for the workshop is required (and limited to 40 people!): Register here.