RAG Interview Questions, issue 9, Jul 13, 2026
The Equal Split Trap
Senior ML Engineer interview at Pinterest, and the interviewer asks:
“You’re building a Product Quantization index. You split each vector into M subvectors and run k-means on each. A teammate says the split point doesn’t matter because the information is spread uniformly. Do you ship it?”
Don’t say: “Sure, embeddings are dense, so any equal split works.”
Why arbitrary chunking tears correlated dimensions apart and quietly ruins vector search, and how rotating your embeddings saves your latency-recall tradeoff.
The full answer, with the mechanism and the arithmetic, is for paid subscribers on Substack.