Data/ML/AIDevOps/CloudAround since 2018
MLflow
Also written as Weights and Biases, Experiment Tracking
A tool that records every model training run — the settings used, the data, and how well the result scored — so a team can tell which version is which.
Think of it like
A logbook for every batch you brew, so when one turns out well you know exactly how you made it.
Junior or senior?
A genuine maturity marker. Teams without it usually cannot reproduce a model they shipped six months ago, and candidates who have felt that pain say so unprompted.
Ask them
“How did your team know which model version was running in production?”