Databricks joins with Tecton in order to accelerate enterprise ML projects after Snowflake

Databricks joins with Tecton in order to accelerate enterprise ML projects after Snowflake ...

Databricks is receiving funding for the Tectons feature store, which is being expanded by an increasing number of businesses looking to leverage their lakehouse platform for machine learning (ML) initiatives.

In a statement on Thursday, Tecton announced an integration that will make its feature store available on Databricks, giving joint customers the opportunity to create and automate their ML feature pipelines from prototype to production in a matter of minutes.

Tecton is expanding its underlying data infrastructure to support ML-specific requirements by leveraging Databricks. This collaboration with Databricks allows organizations to incorporate machine learning into live, customer-facing applications and business processes, quickly, reliably and at an analogue scale, according to Mike Del Balso, the company''s cofounder and CEO.

What is the impact of the Tecton feature store on the deployment of ML applications?

The ML model underneath must be educated on historical data in most instances. This data can be visualized as a table, with rows representing certain elements and columns, which define the elements. Each individual attribute, or measurable property, is a specific feature. The process involves unique engineering challenges and takes a lot of time, affecting training and deployment timelines.

Transform 2022

In person, you can join us for the most important event on applications AI for enterprise business and technology decision makers, from July 19 to August 28.

A feature store allows data scientists to obtain advanced features for reuse at a later stage or by another team member within the same organization. Tecton does the same job, although its offer is ahead and simplifies the whole lifecycle of ML features, from the transformation of raw data to the serving for inference.

Teams can automate the construction of ML features and operationalize ML applications in minutes rather than months, depending on how the whole thing works. Without having to leave the Databricks workspace

All of the features a user builds in Delta Lake are natively available in the data lakehouse. Databricks users also have access to MLflow, where they can host and manage the trained models and produce serving endpoints to deliver real-time predictions. In a simple nutshell, a Databricks user can define and manage features in Tecton, process feature values using Databricks compute, and serve predictions with MLflow.

Widespread adoption

Multiple Tecton and Databricks customers, including Fortune 500 companies, are already using this technology to develop real-time predictive applications such as fraud detection, real-time underwriting, dynamic pricing, and personalization. However, Databricks is not the only company with such an approach.

Snowflake, too, partnered with Tecton to develop its data cloud feature store. The engagement included the integration of its open-source feature store Feast.

You may also like: