Your question is Building Reliable Model Evaluation. 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).
You've trained and shipped a model, and the team wants confidence that its performance will hold up outside offline experiments. You need a clear evaluation approach that catches weak generalization, unstable predictions, and bad decision thresholds before the model causes downstream issues.
How do you ensure that your machine learning models are robust and reliable?