Model Accuracy Drift
A hard SQL mock interview question on DataDriven. Practice with AI-powered feedback, real code execution, and a hire/no-hire decision.
- Domain
- SQL
- Difficulty
- hard
- Seniority
- L5
Interview Prompt
We track model versions and their accuracy over time. For each model, calculate the difference between the accuracy of its most recent version and the average accuracy across all previous versions, rounded to 2 decimal places. If a model has only one version, return 0 for the difference. Show model name, average lifetime accuracy, latest accuracy, and the difference.
Summary
Accuracy used to be higher.
How This Interview Works
- Read the vague prompt (just like a real interview)
- Ask clarifying questions to the AI interviewer
- Write your sql solution with real code execution
- Get instant feedback and a hire/no-hire decision