Here is a debugging session. What is wrong with it?
Answer it out loud before you open anything. The value of the flags below is in comparing them to what you actually said — including whether you asked who uses it before naming what you would build it with.
A production alert: some checkouts in the online store return "Payment failed" although the customer's bank shows a charge. An engineer's terminal history from the last hour is pinned to the incident channel. You are asked to review it and say what to do next.
React to this
Say what you would question, what you would keep, and what you would need to know first.
Incident channel — engineer's terminal history (illustrative): 10:02 kubectl rollout restart deploy/checkout 10:06 redis-cli FLUSHALL 10:09 # bumped PAYMENT_TIMEOUT_MS 5000 -> 15000, redeployed 10:15 kubectl rollout restart deploy/checkout 10:18 git revert HEAD~1 # "maybe the coupon PR" 10:24 # set RETRY_PAYMENT=true, redeployed 10:31 kubectl scale deploy/checkout --replicas=6 10:35 "still seeing some. might be the provider. escalating" Not in the history: an order id, a request id, a log query, a look at the provider dashboard, a database query for a failed order, the error message text.
What it is really testing
Whether the candidate recognises random mutation as the anti-pattern it is, and can replace it with the loop: reproduce, observe, hypothesise, predict, experiment, evidence — and whether they can say which of the engineer's changes destroyed evidence.