Data/ML/AIDevOps/CloudHigh signalEmergingAround since 2017
Feature Store
Also written as Feast, Feature Platform
A shared store of the prepared inputs models use, so the same calculation feeds both model training and the live system.
Think of it like
A single prep kitchen serving every station, so the sauce tastes the same wherever it is plated.
Junior or senior?
Genuinely differentiating, and found only in mature ML teams.
Senior sounds like
Can explain the training-versus-serving mismatch it exists to prevent.
Ask them
“What problem did a feature store solve that your team had before it?”