What’s RAG?
RAG stands for Retrieval Augmented Generation. I like to think of it as using LLMs to try and frame an answer in Natural Language. LLM stands for Large Language Model. Instead of explaining how it works, I think a practical example would benefit more: ChatGPT.
The idea with using RAG is to query an LLM with elevated context which enables it to
Result
You get predictable background execution without surprises.