Top 6 Drift Interview Questions
Browse 6 real-world technical and behavioral interview questions about Drift. Review scenarios, edge cases, and architectural best practices.
What is the difference between data drift and concept drift, and does it change what you do?
Data drift is a change in the distribution of the inputs; concept drift is a change in the relationship between inputs and the outcome. You can detect data drift from inputs alone, concept drift needs labels or a performance signal, and only concept drift necessarily invalidates what the model learned.
How do you decide when to retrain a model?
Retrain when there is evidence the learned relationship no longer holds and you have data that reflects the new one. A fixed schedule is a fallback for when you cannot measure that, not a decision, and a drift alert is as likely to mean an upstream pipeline broke as it is to mean the world moved.
Your labels arrive months after the prediction, or never. What do you monitor?
Monitor what is available immediately: input distributions and data quality, the prediction distribution, and proxy outcomes that correlate with the label but arrive sooner. Do all of it per segment, because a small segment failing badly is invisible in an aggregate.
A model scored well in validation and performs badly in production, with no errors anywhere. What do you check first?
Check whether training and serving compute the same features from the same definitions, because the usual cause is two implementations that disagree slightly. Nothing raises an error when they diverge - the model receives plausible numbers that mean something different from the ones it learned on.
What has to be in place before you would let a model serve live traffic?
Features must be computable at decision time, training and serving must share one transformation implementation, the model and dataset version must be pinned together, drift and realised performance monitored separately because labels arrive late, and rollback must move model and features as one unit.
Why is state the hard part of Terraform, and how do you manage drift, imports and locking around it?
State is the mapping from a config address to a real resource ID, so it is the only thing that tells Terraform whether to create or update. Drift, imports, refactors and concurrent applies are all consequences of that mapping being separate from both the code and the infrastructure.