Rag And Retrieval medium
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?
- A The embedding model drifts over time; re-embed the entire corpus weekly using a fresh model checkpoint.
- B The chunks are too large; smaller chunks would ensure that newer policy text outranks the deprecated text.
- C The generator relies on stale training memory; add a prompt instruction telling it to prefer the newest policies.
- D The job only inserts; have CMS change events upsert changed chunks and delete removed ones by doc ID and version.
Answer: D. The job only inserts; have CMS change events upsert changed chunks and delete removed ones by doc ID and version.
Freshness needs an **invalidation path** from the source of truth to the index. An insert-only job leaves orphaned, outdated chunks that can still be retrieved. Change-driven upserts and deletes add pipeline complexity, but they keep cost proportional to what actually changed. A full re-embed also clears stale content, but it is only justified when you are switching embedding models.
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?