AI Interview DropIntermediate TierMultiple choice+25 XP on read

You have 50M embeddings, need p99 under 50ms, and 95% recall is acceptable. Which vector index do you choose?

Core Summary

Vector index choice is a three-way trade between recall, latency and memory. At tens of millions of vectors with a tight latency budget and tolerance for approximate results, HNSW with quantization is the standard answer: exact search cannot meet the latency, and unquantized HNSW cannot meet the memory budget affordably.

Pick one

Hint 1: 95% recall means you are explicitly allowed to be approximate

Hint 2: Compute the raw memory footprint before choosing

Hint 3: Quantization trades a little recall for a lot of RAM

Reported in interviews at Pinecone, OpenAI

You have 50M embeddings, need p99 under 50ms, and 95% recall is acceptable. Which vector index do you choose?

Core Summary

Hints