I would make the dense vector search the high-recall candidate generator and set its limit to 500. The second stage would use the ColBERT multivector representation to score those candidates more precisely, returning only the final top-k results.

The reason for this architecture is computational economics. Dense ANN retrieval is comparatively cheap and can search a large collection, while late-interaction scoring is more expensive. Restricting ColBERT scoring to 500 candidates gives the reranker a manageable workload while preserving most of the candidate recall.

The important detail is that the ColBERT representation must already be stored as a compatible multivector in Qdrant, and the query must target that vector configuration. The exact query syntax for multivectors and late interaction is version-sensitive, so I would use the API corresponding to the deployed Qdrant release.

A common mistake is setting the prefetch limit equal to the final top-k. That leaves the reranker too few candidates and can cap recall before reranking even starts.

javascript

1
Dense ANN is used for high-recall candidate generation
2
The reranker operates only on the prefetched candidates
3
ColBERT uses a multivector late-interaction representation
4
Candidate count should be tuned for recall, latency, and reranking cost

Difficulty: 9/10

Topics: Multivectors, ColBERT, Reranking, Multi-stage retrieval

Scenario Questions

0-2 years experience

1
Why would a retrieval system use a cheap first-stage search before an expensive reranker?
2
If the relevant document is not among the 500 dense candidates, can the ColBERT reranker recover it? Why not?

2-5 years experience

1
Your reranker is accurate but too slow when given 2,000 candidates. How would you tune the first-stage candidate count?
2
Increasing dense candidates from 500 to 1,000 improves recall slightly but doubles reranking latency. How would you evaluate the trade-off?

5-8 years experience

1
Your ColBERT stage improves precision but creates a p99 latency regression. How would you optimize the two-stage retrieval pipeline?
2
Dense retrieval has strong recall for English but weak recall for multilingual queries. Would increasing the candidate count solve the problem? How would you investigate?

8+ years experience

1
You need to serve a multimodal RAG workload with dense retrieval followed by late interaction under a strict latency SLO. How would you architect and tune candidate generation versus reranking?
2
A new reranker improves NDCG substantially but increases compute cost by 10x. How would you decide where in the retrieval pipeline it belongs?

Follow-up Questions

❓Why is 500 a candidate count rather than a fixed best practice?
❓What happens if the dense prefetch misses a relevant document that ColBERT would have ranked highly?