Companion videos

Building a production RAG — companion videos

The article walks through what changes when a RAG system moves from a notebook to real traffic — ingestion edge cases, embedding choice, hybrid retrieval, rerankers, and the eval harness that holds the whole pipeline accountable. These two AI Engineer Summit talks are the cleanest on-camera versions of that story: one from the LlamaIndex side on what an actual production RAG stack looks like, the other from the original RAG paper's lead author on what has bitten him deploying it across Fortune 500 customers.

Primary pick

18:34

Building Production-Ready RAG Applications: Jerry Liu

AI Engineer

LlamaIndex's CEO walking the gap between "naive RAG demo" and a real pipeline — small-to-big retrieval, sub-question routing, hybrid search, evaluation. The shape of his slides maps almost directly onto the article's pipeline sections; watch first, then re-read the article with his diagrams in your head.

What you should get from this: Identify the production RAG controls missing from naive document-chat demos.

Watch or know first: Have built or evaluated a basic RAG prototype.

AI Expert note: LlamaIndex APIs and recommended components change, but the production gaps are durable: ingestion quality, retrieval routing, reranking, evals and observability. Verify current library defaults before implementing.

Open video page

Also worth watching

16:56

RAG Agents in Prod: 10 Lessons We Learned — Douwe Kiela, creator of RAG

AI Engineer

Douwe Kiela led the original RAG paper at FAIR and now ships RAG into regulated enterprises. The talk is mostly about what stops working at scale — chunking strategies that don't survive 100k documents, "accuracy is table stakes, inaccuracy is the real problem," and why attribution and observability matter more than the embedding model. Good calibration before re-reading the article's eval and monitoring sections.

What you should get from this: Understand what breaks when RAG moves into regulated, high-stakes enterprise use.

Watch or know first: Know the basic RAG pipeline and why attribution matters.

AI Expert note: Strong production cautionary material. Use it to pressure-test architecture decisions, especially permissions, attribution, observability and user trust, before debating embedding-model minutiae.

Open video page