One Bad Executor
A medium AI Coding mock interview question on DataDriven, with an AI interviewer who asks follow-up questions and a hire or no-hire decision at the end.
- Domain
- AI Coding
- Difficulty
- medium
- Seniority
- L5
Interview Prompt
Our nightly orders enrichment job ran 3h 40m last night instead of its usual 22 minutes, with one executor OOMing on stage 4 and cluster spend tripling, even though order volume grew only about 8% week-over-week. Fix the job in jobs/ so tests/test_skew_observability.py passes using only what the production cluster will actually accept, and be ready to defend your trade-offs against a different data shape.
Summary
One node held the whole night hostage.
How This Interview Works
- Read the task and the repository: the files, the failing check and the constraints
- Work the fix with the AI agent at your side, and run the checks until they pass
- Submit, then answer the interviewer's questions about the change you made
- Get a hire or no-hire decision with feedback on how you drove the agent