Your question is Chunking and Vector Indexing. 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 indexing strategies for effective vector search?
Explain how you would split documents, preserve useful context and metadata, generate embeddings, and select an index for retrieval. Include a practical Python implementation and describe how you would evaluate chunk quality, retrieval relevance, latency, and update behavior.