Snowflake to accelerate ML projects with Tecton and Feast feature stores

Snowflake is getting new feature stores, as an increasing number of enterprise teams look at the data company to build and deploy machine learning applications.

In a statement on Wednesday, Tecton announced a partnership with the data giant under which the former’s feature store, known for managing the complete lifecycle of machine learning (ML) features, as well as the open-source one from Feast will be integrated with the Snowflake Data Cloud. The move, as the companies explained, will give enterprise data scientists a fast yet simple way to build production-grade features for a broad range of operational ML use cases, starting from fraud detection and product recommendation to real-time price tracking.

Problems with ML projects

Enterprises using cloud data platforms (such as Snowflake) for ML projects can run into issues such as distinct pipelines during implementation or training data leakages/inaccuracies. This can slow the development time, affecting the delivery of the project. There’s also no way for enterprise users to discover and re-use features that have already been created by other data scientists.

With the integration of Tecton and Feast, Snowflake users can easily address these challenges. Tecton, a central hub for ML features, allows data teams to define features as code using Python and SQL and then automates production-grade ML data pipelines, generates accurate training datasets and serves features online for real-time inference. 

“It allows organizations to “just turn on” real-time capabilities once they want to move into real-time ML. And it provides a central catalog of features that can be discovered and re-used by other teams, significantly speeding up development as ML organizations mature,” Mike Del Balso, cofounder and CEO of Tecton, told Venturebeat. 

Similarly, Feast — accessible through a Snowflake connector — also acts as an interface to operationalize analytic data for model training and online inference. The open-source feature store counts thousands of active users and is already integrated with Redshift, BigQuery, Databricks and S3. Meanwhile, Tecton includes companies like Atlassian, Tide, and Fortune 500 insurance players in its paying customer base and comes integrated with Databricks and S3 only.

Snowflake’s data science play 

The move comes as the latest step from Snowflake to strengthen its data science play – one of the six workloads its supports through the Data Cloud along with data lake, data warehouse, data engineering, data application and data sharing.

“Together with our partners, Snowflake will continue innovating to improve user experience at every step of the machine learning workflow, including feature engineering, model training, and model deployment. Snowpark for Python, currently in private preview, is a huge step forward,” said Julian Forero, senior product marketing manager at Snowflake. “It enables teams to collaborate on a single copy of governed data, use their preferred programming language, and access a rich ecosystem of open-source libraries while taking advantage of Snowflake’s elastic performance engine.”

Prior to this, the company had announced the acquisition of Streamlit to strengthen the data application side of its platform. The deal was reportedly closed at $800 million.