Platform Component: Model Ops

Deploy and Manage Models to Drive Business Impact

Domino’s Model Ops capabilities allow you to export your production models to the infrastructure of your choice. You can see all of your production assets – APIs, apps, scheduled jobs, and more – at a glance to understand how they’re impacting business.

With the optional Domino Model Monitor, you can standardize model monitoring and view the health of all models in a single pane of glass to ensure that critical business decisions are being made with the most accurate and up-to-date models.

Deploy Data Science Models on Any Infrastructure

Deploy Data Science Models on Any Infrastructure

  • Turn your models into on-demand REST APIs with high availability, horizontal scalability, and security
  • Export models as Docker images to CI/CD pipelines, AWS SageMaker, or other infrastructure
  • Add CI/CD pipelines and other guardrails to ensure models get deployed only when they’re ready
  • WEBINAR: Data-driven to Model-driven
Publish Powerful Data Science Applications for Business Users

Publish Powerful Data Science Applications for Business Users

  • Create interactive apps with tools such as Shiny, Dash, and Flask to perform "What-If" analysis and review different scenarios
  • Run templatized analyses using lightweight web forms to let non-technical users interact with the underlying model
  • See usage statistics and trends to understand utilization and the overall business impact
  • WEBINAR: Best Practices for Getting Data Science Web Apps in Production
Proactively Monitor Models to Maximize Business Impact

Proactively Monitor Models to Maximize Business Impact

  • Detect drift in input features and output predictions using advanced statistical checks
  • Register ground truth to detect performance drift
  • Schedule automated checks and alerts to save data scientists' time and rapidly respond to model concerns
  • Explore data using our drift analyzer UI to diagnose drift quickly
  • WHITE PAPER: Model Management: A Framework to Build a Model-Driven Business

The testers in the field sign in through Domino and can play with the model to see if it would help them with their claims processing. As they gave us feedback, we were able to roll that feedback into the app overnight. We got to test it really quickly and got to that learning a lot quicker.

Stephen Collins

Stephen Collins

Data Science Manager

Allstate

Domino Data Lab Enterprise MLOps Platform | Customer stories

Unleash data science

See why over 20% of the Fortune 100 has chosen Domino

Dun & Bradstreet seal