DataRobot announced the latest integration with Snowflake. Building off of DataRobot’s expanded partnership and existing integration with Snowflake, the new Feature Discovery pushdown integration improves the speed and accuracy of developing models, unlocking new use cases. DataRobot’s Feature Discovery, which has been a part of the DataRobot enterprise AI platform since 2019, automatically discovers, tests, and creates hundreds of valuable new features for machine learning models. This improves models’ accuracy, increasing an organization’s ability to make accurate predictions.

The new Feature Discovery integration with Snowflake delivers this capability to Snowflake users, pushing down data preparation operations into Snowflake to minimize data movement resulting in faster performance and lower operating costs. This allows users to obtain more accurate DataRobot models by accessing more data from Snowflake and leveraging the power of Snowflake’s Data Cloud. With Feature Discovery, the joining, aggregating, and creation of derived features from datasets is done automatically using data science best practices. This lets users build better machine learning models in less time and drive more innovation with AI.