The average doctor sees around one hundred thousand patients in their lifetime. In this talk, I’ll show you how you can build a super-doctor capable of retrieving from over 150,000+ unique patient cases and 3M medical publications to answer patient-related questions. I will use advanced retrieval augmented generation(RAG) techniques such as hybrid search, re-ranking, query generation, and semantic text chunking. Combine these with some of the best LLMs around, specialized biomedical embedding models, and scalable vector databases and we can build a medical chatbot with more information than any doctor alive!