# Feature Store

> A feature store is a system that manages the input variables ("features") used by machine-learning models, ensuring the same, consistent values are available for both training and live prediction.

*Source: https://www.lazlosoftwaresolution.com/glossary/feature-store*

Feature stores solve training/serving skew — the bug where a model behaves differently in production because features were computed inconsistently. They centralise, version and serve features so models stay reliable.