Skip to content
SPECIFICTION
Get listed

Feathr

github.com

WebLinuxAPISelf-hosted
Visit github.com
Spec score 7.6 #5 of 20 in Feature Store Software

What it is

Feathr is a free, open-source platform for building and sharing data features used in machine-learning workflows. Its Python client supports custom functions alongside PySpark and Spark SQL, and a unified API runs transformations in batch, streaming, or online settings. Feature operations include time-based aggregations, sliding-window joins, lookups, and derived features. Feathr joins computed features to training data with point-in-time correctness to help reduce data leakage. An optional registry and UI support feature search, metadata and lineage review, sharing, and access management; registry backends include Azure Purview and ANSI SQL. Integrations include Azure and AWS storage, Snowflake, Kafka, Redis, Databricks, and Azure Synapse, as well as Azure Machine Learning and notebook tools. The project documents Azure, Databricks, and Synapse deployment, a Docker sandbox, and a locally installable Python client. It runs on Linux and web, and supports self-hosted and API use. The repository specifies Apache License 2.0. The FAQ notes that preprocessing currently appears limited to one UDF function, and feature-level access control is not supported yet.

Spec sheet

Purpose
Feathr is a data and AI engineering platform for defining, registering, and sharing data and feature transformations.Read 4 Oct 2026 · github.com
Deployment
The project documents Azure deployment and provides a self contained Docker sandbox, plus a locally installable Python client.Read 4 Oct 2026 · github.com
API
The registry deployment exposes a REST API, and both the Feathr UI and Python client interact with it.Read 4 Oct 2026 · feathr-ai.github.io
Integrations
Listed integrations include Azure Blob Storage, ADLS Gen2, AWS S3, Azure SQL, Snowflake, Kafka, EventHub, Redis, Azure Cosmos DB, Databricks, and Azure Synapse.Read 4 Oct 2026 · github.com
Support
The project directs users to its Slack channel for questions and discussions.Read 4 Oct 2026 · github.com
Founded
2017Read 28 Sep 2026 · github.com
AI modeling
Feathr computes feature transformations and joins them to training data using point-in-time-correct semantics to help avoid data leakage.Read 3 Oct 2026 · github.com
Cloud and compute integrations
Documented integrations include Azure Synapse, Databricks, Azure Machine Learning, and Jupyter Notebook.Read 3 Oct 2026 · github.com
Community support
The project directs users to its Slack channel and GitHub Discussions for questions and discussion.Read 3 Oct 2026 · github.com
Execution modes
Its unified data transformation API works in offline batch, streaming, and online environments.Read 4 Oct 2026 · github.com
Feature engineering
It supports time based aggregations, sliding window joins, lookup features, and derived features with point in time correctness.Read 4 Oct 2026 · github.com
Feature registry
The built-in registry supports searching features, viewing data sources and lineage, and managing access controls.Read 3 Oct 2026 · github.com
Installation and deployment
The project documents installing its Python client with pip and deploying Feathr on Azure, Databricks, or Azure Synapse; its sandbox is distributed as a Docker container.Read 3 Oct 2026 · github.com
Intended use
The FAQ says a feature store is typically useful when modeling entities such as users, accounts, or items, and may not be necessary for regular image recognition.Read 4 Oct 2026 · feathr-ai.github.io
Interfaces
Feathr provides Pythonic APIs and customizable user defined functions with native PySpark and Spark SQL support.Read 4 Oct 2026 · github.com
License and availability
The GitHub repository is public and its license file specifies Apache License 2.0.Read 3 Oct 2026 · github.com
ML tools
The project lists Azure Machine Learning, Jupyter Notebook, and Databricks Notebook as machine learning platform integrations.Read 4 Oct 2026 · github.com
Notable limit
The project FAQ says preprocessing currently seems to accept only one UDF function, subject to change based on requirements.Read 4 Oct 2026 · feathr-ai.github.io
Processing modes
Its unified transformation API supports offline batch, streaming, and online environments.Read 3 Oct 2026 · github.com
Project status
The README says Feathr was open sourced in 2022 and is a project under the LF AI & Data Foundation.Read 4 Oct 2026 · github.com
Project stewardship
The repository says Feathr is a project under the LF AI & Data Foundation and was open sourced in 2022.Read 3 Oct 2026 · github.com
Registry and governance
The optional UI and registry let users search features, inspect metadata and lineage, manage access controls, and share features across teams.Read 4 Oct 2026 · feathr-ai.github.io
Registry backends
The feature registry supports Azure Purview and ANSI SQL backends; role based access control requires a SQL service to store its related information.Read 4 Oct 2026 · feathr-ai.github.io
Scale
The project says Feathr can process billions of rows and petabyte scale data using optimizations such as bloom filters and salted joins.Read 4 Oct 2026 · github.com
Secret handling
The configuration guide says settings can be stored in Kubernetes secrets or a key vault, and that Azure Key Vault is currently supported for retrieving values.Read 3 Oct 2026 · feathr-ai.github.io
Security controls
The registry access control documentation describes project-level role-based access control with admin, producer, and consumer roles, and says feature-level access control is not supported yet.Read 3 Oct 2026 · feathr-ai.github.io
Storage and streaming integrations
Documented options include Azure Blob Storage, Azure ADLS Gen2, AWS S3, Snowflake, Kafka, EventHub, Redis, and Azure Cosmos DB.Read 3 Oct 2026 · github.com
Transformation API
Feathr provides Pythonic APIs and customizable user-defined functions with native PySpark and Spark SQL support.Read 3 Oct 2026 · github.com

Strong points and limits

Strong points

  • Point-in-time joins help avoid training-data leakage.
  • Supports batch, streaming, and online transformations.
  • Registry provides feature search, metadata, and lineage.
  • Apache License 2.0; available free.

Limits

  • Feature-level access control is not supported yet.
  • Preprocessing currently seems limited to one UDF.
  • Feature monitoring is not supported.

Who it suits

Feathr suits teams that build machine-learning features across batch, streaming, or online workflows and need a shared registry. Its feature-store approach may be less relevant to regular image recognition, according to the project FAQ.

Verdict

Feathr provides a broad feature-engineering workflow, from transformations to registry and deployment options. Check its access-control and preprocessing limits against your requirements before adopting it.

Where Specifiction places it

Alternatives to FeathrAll alternatives →