Your question is Real-Time Anomaly Detection at Scale. 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).
Design a real-time anomaly detection pipeline that ingests millions of spans, metrics, and logs per second.
Clarify detection latency, alert precision and recall, retention, tenant isolation, and cost requirements. Describe the end-to-end ingestion, feature computation, anomaly scoring, alerting, feedback, training, and monitoring architecture. Explain how the design handles seasonality, new services, noisy telemetry, delayed labels, feature drift, training-serving skew, and failures at scale.