This endpoint used to answer in about 100 ms. Since Tuesday, p99 is around 3 s and p50 has barely moved. Where do you start?

Answer it out loud before you open anything. The value of the flags below is in comparing them to what you actually said.

The situation behind the question

A checkout listing endpoint in a service that has not been rewritten, only deployed to. Error rate is unchanged; nobody is paging, but support tickets mention "the page hangs sometimes". Tuesday had four deploys and a marketing email.

What it is really testing

Whether the candidate reasons from the *shape* of the regression — p99 moved and p50 did not — or immediately proposes a fix. A tail-only regression means most requests are fine and a minority are waiting on something finite or slow; that rules out a large class of causes before any code is read.

Where the mechanism is taught