Virtualization Technology News and Information
Tecton Becomes Feast Core Contributor to Build the Most Advanced Open Source Feature Store for Machine Learning
Tecton announced that it will become a core contributor to Feast and allocate engineering and financial resources to the project to build advanced capabilities. Feast is the leading open source feature store for machine learning (ML) that bridges data and models and allows ML teams to deploy features to production quickly and reliably.

Feature stores are emerging as a critical component of the infrastructure stack for operational ML. Tecton will continue to advance its production-ready enterprise feature store that is delivered as a fully-managed cloud service and is trusted by some of the world's biggest brands. Tecton's contributions to Feast will offer users the freedom to choose between open source software and commercial software.

As part of today's announcement, Tecton announced that the creator of Feast, Willem Pienaar, will be joining the company. He will remain fully-committed to Feast and an official maintainer of the project; Feast will continue to be an independent open source project managed by the LF AI & Data Foundation, part of the Linux Foundation. 

"Inspired by the Michelangelo ML platform, we created Feast to accelerate Gojek's deployment of ML-powered applications," said Willem Pienaar, creator of Feast and architect at Tecton. "We open sourced the project in response to broad community interest and are delighted to see strong user adoption. Collaborating with Tecton will allow us to accelerate our roadmap and build the best open source feature store for the community."

Feast was developed jointly by Gojek and Google Cloud. Since its initial release in 2019, Feast has grown rapidly, with multiple companies, including Microsoft, Agoda, Farfetch, Postmates and Zulily adopting and/or contributing to the project. The project has more than 1,100 GitHub stars.

"Operationalizing data is the hardest part of getting ML to production," said Matt Ziegler, lead software engineer at online retailer Zulily, a contributor to Feast. "The Feast feature store allows our team to bring DevOps-like practices to our feature lifecycle. Data scientists now have a single source of truth for data and can quickly serve feature values for training and online inference, enabling us to further personalize shopping experiences. It's great to see Tecton supporting Feast, adding cross-industry expertise to the project and further building an interface between data and models in production."

Tecton and Feast will build a simple migration path that will give users the freedom to transition between Feast open source software and the Tecton feature store. Tecton and Feast will have fully compatible serving APIs to make the migration transparent to models and applications.

"We're excited to be core contributors to the Feast open source project," said Mike Del Balso, co-founder and CEO of Tecton. "Feature stores allow ML teams to build and deploy new features to production within hours instead of months. Users will benefit from the freedom to choose between open source software and Tecton with the flexibility to migrate transparently between the two."

Tecton's enterprise feature store serves as a central hub for the data processes that power operational ML models. It allows data scientists and ML teams to build features and deploy them to production in a fraction of the time. Tecton is delivered as a fully-managed cloud service with guaranteed service levels and enterprise support.

Published Monday, November 16, 2020 3:36 PM by David Marshall
Filed under:
There are no comments for this post.
To post a comment, you must be a registered user. Registration is free and easy! Sign up now!
<November 2020>