Your question is Embedding Model Trade-offs. 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).
What are the primary trade-offs between various embedding models when building a domain-specific vector search engine?
Explain how you would compare general-purpose, domain-pretrained, and fine-tuned embedding models in a practical retrieval pipeline. Address semantic quality, vocabulary coverage, dimensionality, indexing cost, inference latency, multilingual behavior, and maintenance requirements. Provide a small Python evaluation approach using representative queries and documents, and describe how you would select a production model based on retrieval metrics and operational constraints.