Your question is Retrain on Data Drift Pipeline. 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 would you architect a pipeline to retrain models based on incoming data drift at Datadog?
Explain how you would detect meaningful drift, distinguish it from expected traffic or seasonal changes, trigger retraining, and validate a new model before deployment. Cover data quality, feature drift, training-serving skew, orchestration, rollback, monitoring, and human approval points. State which assumptions you would clarify about model criticality, data volume, retraining frequency, latency, and compliance.