Redis Feature Form

About Redis Feature Form

Redis Feature Form is an enterprise feature store platform for production machine learning. It helps teams define ML features as code and convert those definitions into production grade data pipelines for model training and inference.

The platform orchestrates between offline data systems and real time serving. It works with existing data stacks such as Snowflake, Databricks, Spark, and Postgres while using Redis as the online store for low latency feature retrieval.

Redis Feature Form is designed for enterprise AI and ML teams that need reusable features, governed pipelines, workspace isolation, versioning, lineage, observability, and secure access controls across multiple teams.

Key Features

• Feature Definitions as Code: Define features once and use them for both model training and inference.
• Real Time Serving: Use Redis as an online store for low latency feature retrieval in production.
• Pipeline Orchestration: Turn feature definitions into computation across offline data systems and real time stores.
• Workspace Isolation: Run multiple ML teams with scoped access, isolated providers, and independent observability.
• Versioning and Lineage: Track how features evolve and understand which models depend on them.
• Enterprise Security: Use workspace scoped RBAC, API key pairs, audit logs, mTLS, encrypted transport, and secret providers.