Explore the intricacies of Retrieval Augmented Generation (RAG) in this comprehensive conference talk presented by Greg Loughnane and Chris Alexiuk from AI Makerspace. Delve into the components of a simple RAG system, including vector stores, embedding models, and LLMs, and learn how they are integrated using LLM Ops infrastructure. Gain hands-on experience with LangChain tooling to embed documents using a leading embedding model, store them in a Pinecone vector database, and pass augmented prompts to Llama 2. Discover evaluation techniques and emerging best practices for optimizing RAG outputs. Access all demo code via GitHub and/or Colab to implement your own RAG system efficiently.