Your question is Handle Model Drift and Retraining. 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 model drift, anomaly detection, and automated retraining pipelines in production at Netflix?
Explain an end-to-end production design covering data and prediction monitoring, drift and anomaly detection, retraining triggers, validation, deployment, rollback, and human oversight. Address delayed labels, training-serving skew, model versioning, experiment safety, and how you distinguish data-quality incidents from genuine behavioral change.