Rag And Retrieval medium
Runbooks are split into fixed 300-token chunks with no overlap. When a procedure step starts near the end of one chunk and finishes in the next, answers come back incomplete. What is the lowest-cost fix?
- A Switch to 3,000-token chunks so that no runbook step can ever be split across two separate chunks.
- B Add 10-20% overlap between adjacent chunks, or split on step and heading boundaries instead of token counts.
- C Raise top-k from 5 to 40 so that both halves of every split step are likely to be retrieved together.
- D Add a reranker so the chunk with the first half of the step is ranked above unrelated chunks in the list.
Answer: B. Add 10-20% overlap between adjacent chunks, or split on step and heading boundaries instead of token counts.
**Overlap** and **structure-aware splitting** protect against boundary loss. The cost is roughly 10-20% more embeddings and storage, plus some duplicated context. Very large chunks also avoid splits, but they dilute relevance and raise prompt cost. A higher k does not guarantee both halves of a step rank highly. Fixed chunks without overlap are acceptable only for content made of short, self-contained entries.
One system design question a day, with spaced repetition so the ones you miss come back.
Get the appRelated questions
- You must index 100M chunks using a 3,072-dimension embedding in fp32, served from RAM via HNSW. The raw vectors alone take about 1.2 TB before graph overhead. You need to cut memory several-fold while losing at most a point or two of recall@10. Which approach fits?
- A multi-tenant SaaS app stores all tenants in one vector index with a tenant_id field. Queries fetch the top 20 chunks by similarity and then drop chunks from other tenants. Small tenants often get 0-2 results. Why does this happen, and what is the fix?
- Your offline eval shows recall@5 of 92% on a labeled set, so the gold passage is in the context for nearly every question. Yet human graders mark 35% of answers wrong, and the wrong answers often ignore or contradict the gold passage. Where should the team invest next?
- Your product requires every RAG answer to include clickable source citations that users can verify. How should those citations be produced?
- A runbook RAG system uses fixed 500-token chunks with no overlap. Users who ask for a multi-step recovery procedure get answers that stop at step 4 of 9, and retrieval logs show the top hit ends mid-procedure. What is the most targeted fix?
- Policy documents live in a CMS. A nightly job embeds and inserts new documents into the vector index. Users keep getting answers that cite policies which were deleted or replaced weeks ago. What is the root cause and fix?