The model does not know your data. Give it eyes.
L6 · Grounding Models
Retrieval is a search engine bolted to a writer. Get the search right and the writing mostly takes care of itself, which is why this chapter is about chunking, indexes and ranking rather than about prompts. You will rank passages by meaning, cut documents in the place that keeps a rule whole, measure what an approximate index gives away, and merge a word search with a vector one. Then you will separate retrieval failures from writing failures, which is the skill the rest of it rests on, and work out whether any of it beats pasting the whole handbook into the window.
- Why the Model Does Not Know Your Data
- Search by Meaning
- Chunking Decides Everything
- Vector Indexes
- Keywords Still Matter
- Read the Shortlist Properly
- Answers With Receipts
- Retrieval in Arabic
- Measuring the Two Halves
- Four Ways It Fails
- Paste It All, or Go and Find It
- Teach It, or Show It
- Every Tool Is a Retriever