Your question is Embedding Chunking and Retrieval. 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 do you handle document chunking and retrieval in a system using embeddings and vector search?
Explain how you would split documents into useful passages, generate embeddings, index them, and retrieve relevant context for a downstream language model. Cover chunk size and overlap, metadata, filtering, top-k selection, hybrid or reranked retrieval, and how you would evaluate retrieval quality. Include a practical Python implementation using modern NLP libraries and discuss common failure modes such as lost context, duplicate chunks, and irrelevant high-similarity matches.
