Rag And Retrieval easy
Your RAG index covers 2 million product documents, and about 1,000 of them change each day. A nightly full re-embedding job takes 10 hours. What is the better design?
- A Keep the full rebuild but run it weekly instead, since most documents rarely change at all.
- B Upsert only changed docs' chunks on change events, keyed by document ID and version.
- C Skip re-indexing entirely and let the LLM correct outdated details from its training data.
- D Append new chunks without deleting old ones, and rely on the reranker to prefer the newest.
Answer: B. Upsert only changed docs' chunks on change events, keyed by document ID and version.
**Incremental upsert with versioning** updates only the 0.05% that changed, giving near-real-time freshness at a fraction of the compute, at the cost of building a change-event pipeline. Option A cuts compute but makes data staler, and option D leaves outdated chunks competing with current ones. A full re-embed is only required when the embedding model or chunking strategy changes.
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?