End-to-end model development, deployment, & monitoring

Domino natively integrates with Snowflake - with credentials or OAuth. With just a few clicks, data science teams have immediate access to data residing in Snowflake - without having to orchestrate the movement of data through manual workarounds.

From Domino, Data scientists can build and train models using Snowflake Snowpark, using Python-based libraries in Snowflake’s compute environments where the data resides.

Combine the flexibility of model building in Domino with the scalability and power of Snowflake for in-database computation. 

User-defined functions (UDFs) for ML inference built in Domino are optimized to execute asynchronously on Snowflake, so data scientists can execute Python scoring code directly in Snowflake - where the data resides - to improve security and prediction response time for mission-critical applications.

A single pane of glass shows all exported models with performance indicators, hosting settings, and metadata.

Automatically configure models for drift and model quality analysis. With a few easy steps, set up prediction data capture pipelines and ground truth datasets.

Configure drift tests and thresholds on a  per-feature basis, with automated user notifications - all in Domino’s simple interface. Automatically identify when new production data is available, and update data drift and model calculations. 

Enforce best practices for optimal model performance and better decision-making.