Top 4 Tracing Interview Questions
Browse 4 real-world technical and behavioral interview questions about Tracing. Review scenarios, edge cases, and architectural best practices.
Your intern interview is in a shared document with no compiler. How do you write code you cannot run?
Coding without a compiler means doing the computer's checking yourself. Write two or three test cases before the code, keep functions small with clear names, then trace one test by hand line by line, aloud. The trace finds the bugs a run would have found, and it shows the interviewer how you verify.
Engineers added a user_id label to a few metrics and the observability bill went up tenfold. Explain what actually happened, and how you would fix it without losing the ability to debug.
Every distinct combination of label values is a separate time series with its own index entry, so an unbounded label multiplies series count by the number of users rather than adding one dimension. Metrics are the wrong tool for per-entity questions: keep labels bounded, move per-user debugging to exemplar-linked traces, and guard against the next unbounded label.
Latency on a checkout endpoint has tripled and nobody knows why. What should your metrics, logs and traces already have told you before anyone opens an editor?
Metrics tell you something is wrong and when it started, traces tell you which hop owns the latency, logs tell you what happened to one request. Cardinality is what makes metrics expensive, sampling is what makes traces affordable, and the goal is answering new questions from data already held.
What do you instrument and alert on for an LLM feature running in production?
Trace every request with prompt and model versions, token counts and cost, then add sampled groundedness grading, refusal and retry rates, validation failures and implicit user signals. Alert on countable rates, because an averaged quality score moves too little and too late to page anyone.