Lesson 05 — the decision a retriever can never undo. Most teams discover this only after a wrong citation goes out to a regulator. Before retrieval, a chunker has already decided what’s findable A RAG system is usually drawn as a pipeline: documents come in on the left, an answer comes out on the right.…

Lesson 06 — why most RAG production systems run two retrievers, not one. Pick the wrong one and the model never sees the right document. BM25 and dense fail on opposite queries BM25 is the lexical retriever. It scores documents by how often the query’s tokens appear, adjusted for document length and rare-term weight. It…

Lesson 07 — hybrid retrieval gets the right chunk into the top twenty. Reranking gets it into the top three. The cost is real, measurable, and almost always worth paying for regulatory work. The bi-encoder is a lookup; the cross-encoder is a reading Hybrid retrieval (lesson 06) hands the prompt builder a ranked list. The…

Lesson 08 — the chunk that scored highest is not always the chunk the LLM needs. Retrieval precision and serving scope are two decisions, not one. One chunking pass makes the same slice of text play five roles The default RAG pipeline does one chunking pass. Whatever you chunked is what you embed, what you…

Lesson 09 — when a RAG system gives a wrong answer, the architect’s first job is to name what failed. Three modes, three fixes, one diagnostic vocabulary. Why diagnosis matters more than blame Every wrong RAG answer prompts the same conversation. The product manager says the model is bad. The engineer says the prompt needs…