Welcome to your interview.
The question is on your right: Detect Production Drift in Models. Take a moment with it first.
Talk your thinking through with me if you like - when you're confident, submit your answer and I'll grade it like a real screen (7/10 or better passes). Discussion and graded submissions share your five interviewer interactions, so spend them well.
You've shipped a model that looked stable in offline testing, but production performance has started to move. The team wants to know whether the issue is a change in the input data, a change in the target relationship, or both.
How do you detect data drift and concept drift in production?