How do you choose an embedding model, and how do you handle embedding drift when you upgrade it?

The embedding model is the foundation of retrieval, and swapping it is deceptively dangerous. The signal is choosing on domain-relevant retrieval quality (not leaderboard) and knowing that a new model means re-embedding everything. Here is the answer.

Updated Aug 2026 · Grounded in real Applied AI Engineer interview loops and written to a senior-engineer editorial bar.