there are TONS of tools and tutorials out there to build LLM pipelines with RAG, but somehow we STILL struggle to successfully build, deploy, and maintain our custom RAG systems over time 😔
We (our team at uc berkeley) want to understand why—what do we not expect to be hard, and how can we solve them? if you are an engineer building RAG systems and have a free hour to chat send me a DM!
We are looking to talk to people building RAG systems to see how --- past the tutorials --- they struggle when building production RAG systems! What debugging processes are developers going through? What are their main bottle necks? etc... If you know anyone who might be interested, feel free to send me a DM!