Rag And Retrieval easy
A customer deletes a confidential document from your SaaS app, but the RAG assistant still quotes it a week later. What design gap most likely caused this?
- A Deletes in the source never propagated to the index, so the doc's chunks stayed searchable.
- B The embedding model memorized the document during indexing and keeps reproducing its text.
- C The reranker cached its scores, so it keeps promoting the deleted document's chunks.
- D Chunk overlap copied the document's text into neighboring chunks that belong to other documents.
Answer: A. Deletes in the source never propagated to the index, so the doc's chunks stayed searchable.
**Freshness needs an invalidation path** from the source of truth to the index; without delete events, removed content stays retrievable. Building that path costs pipeline work, but it is mandatory for correctness and compliance. Option B is tempting, but embedding models only map text to vectors and do not memorize the corpus; the text comes back because the chunk is still stored.
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?