Your question is Isolating a Bottleneck in an Enterprise System. 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).
Walk me through how you would isolate a bottleneck in an enterprise system.
Explain how you would form and test hypotheses using system telemetry, dependencies, traces, logs, and resource metrics. Cover where ML could assist with anomaly detection or root-cause ranking, how you would distinguish infrastructure, database, network, and application bottlenecks, and how you would validate remediation without introducing new failure modes.