AppliedAIPrep logoAppliedAI/Prep
← MLOps & ML Engineering / 48
mediumGoogleAmazonDatabricks

Walk me through concept drift, data drift, and label drift. Which one actually forces a retrain?

Three drifts get casually lumped together, but they have different detectability and different fixes. The interview signal is knowing which you can catch without labels and which one genuinely demands a new model.

Updated Aug 2026 · Grounded in real Applied AI Engineer interview loops and written to a senior-engineer editorial bar.

Three drifts get casually lumped together, but they have different detectability and different fixes. The interview signal is knowing which you can catch without labels and which one genuinely demands a new model.

Unlock the other 750 answers · ₹2,000 / $25Your progress and mastery stay saved · 6 months · one payment · no auto-renew
UP NEXT ON YOUR JOURNEY
Next in this trackHow do you design the triggers and cadence for retraining a fleet of production models?Next in this trackLabels arrive weeks late. How do you monitor a model from its predictions and outputs alone?Next in this trackHow do you define SLOs and error budgets for an ML system, where 'correct' is probabilistic?
DISCUSSION · 0

No comments yet — be the first to share your approach.