Data/ML/AIDevOps/Cloud
Model Registry
Also written as Model Versioning
The catalogue of trained models, tracking which version is approved, which is live, and how to roll back to the previous one.
Think of it like
A library where every edition is shelved and dated, so you can always find the one you were reading last month.
Junior or senior?
Pairs with experiment tracking and model serving. Its presence suggests a team that ships models repeatedly, rather than one that launched a model once and left it.
Ask them
“How did you roll back a model that was performing badly in production?”