Your question is Real-Time Predictive Analytics Design. 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 predictive analytics system that can handle high-throughput data streams.
Clarify the prediction objective, event schema, freshness requirements, throughput, latency, and availability before proposing an architecture. Cover ingestion, stream processing, feature computation, model training, online inference, feedback collection, and operational safeguards. Explain how you would choose between batch and online processing, support model and feature updates, evaluate predictions with delayed labels, and handle failures such as backpressure, feature drift, training-serving skew, duplicate events, and model-serving outages. State the scale assumptions you need from the interviewer rather than treating unspecified traffic or catalog values as known.