Your question is Embeddings for Semantic Search at Scale. Take a moment with it on the right.
Talk me through your thinking if you like. When you're confident, submit your answer and I'll grade it like a real screen (7/10 or better passes).
How would you build and serve embeddings for semantic search at scale?
Explain how you would clarify traffic, corpus size, freshness, latency, availability, and cost requirements before selecting models and infrastructure. Cover offline embedding generation, vector indexing, online query inference, retrieval, monitoring, evaluation, and failure handling, including training-serving skew and embedding drift.