Rag And Retrieval easy
Your chunker splits policy documents every 400 tokens with no overlap. Users report that answers miss conditions such as 'except for contractors' that appear just after a split point. What is the cheapest fix?
- A Add 10-20% overlap between adjacent chunks so sentences near boundaries appear in both of them.
- B Increase chunk size to 2,000 tokens so that splits happen far less often across each document.
- C Raise top-k from 5 to 50 so the neighboring chunk is also more likely to be retrieved.
- D Switch to a higher-dimension embedding model so each vector captures more of the surrounding text.
Answer: A. Add 10-20% overlap between adjacent chunks so sentences near boundaries appear in both of them.
**Overlap guards against boundary loss** at a small cost: some duplicated storage and a few repeated tokens in context. Option B removes some boundaries but dilutes relevance and inflates context cost, and option C adds noise without guaranteeing the neighbor is retrieved. If documents have clear headings, structure-aware splitting on section boundaries can beat fixed-size overlap.
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?
- 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?
- 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?