feature store

Definitions

1. A centralized repository for storing, managing, and serving machine learning features — the computed input variables used by ML models. Feature stores ensure consistency between training and serving (avoiding training-serving skew), enable feature reuse across teams, and handle both batch and real-time feature computation.

   In plain English: A shared library of pre-computed data inputs for AI models, so every team uses the same calculations and the numbers in training match what the model sees in production.

   Example: Instead of each team re-computing 'user_avg_session_duration' differently, they pull it from the feature store where it is defined once and served with sub-10ms latency.

Related Terms

• feature engineering
• model serving
• MLOps