Your question is Data Consistency and Latency. 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 data consistency and latency in a distributed system?
Explain the architectural decisions, consistency model, replication approach, caching strategy, and failure handling you would use. Address how these choices affect ML feature freshness, online inference, observability, and recovery.