Increase Model Velocity
Domino's integrated model factory removes friction from all phases of the end-to-end data science lifecycle. Rapidly test ideas, select the best models, and deploy them into production to deliver business value.
Getting data scientists quickly to work on solving business problems is key for scaling data science and driving business impact. Domino’s data science workbench provides the flexibility and power that data scientists need to accelerate research. They are free to use the tools they want, on hardware optimized for the task at hand, in a governed and scalable environment that fosters reproducibility, reusability, and collaboration.
Domino lets you run data science and machine learning workloads across any compute cluster — in any cloud, region, or on-premises. MLFlow integration makes it easy for data science teams to run hundreds of machine learning experiments in parallel and easily compare the results, maximizing their productivity.
Integration with Feast provides a single source of truth for features in the organization, driving reuse, consistency, and reproducibility.
Model and App Deployment
Easily export production models to infrastructure of choice
Business value doesn’t accrue until you get models into production. Without repeatable processes and automation to expedite the validation and deployment of models, organizations at best delay ROI and often never see the full potential of their data science work.
With Domino, you can accelerate processes to create, manage, scale, and secure production models. Models are easily deployed through scalable APIs, Apps, and Launchers, or exported as Docker images to CI/CD pipelines, AWS Sagemaker, or other infrastructure. Interactive, scalable Apps created with Shiny, Dash, and Flask make it easy for non-technical users to interact with models.
Integrated Model Monitoring
Automatically track data drift, model quality and other health statistics
Models left to their own devices can quickly derail a business. But many organizations struggle to effectively monitor models in production and efficiently remediate issues to keep models at peak performance.
Domino's integrated model monitoring provides a “single pane of glass” for observing traffic, drift, and health trends for all production models with out-of-the-box and custom metrics and KPIs. You will be automatically alerted when drift, divergence, and data quality checks exceed thresholds. When retraining is needed, it's easy to drill down to model features to modify, retrain and redeploy models quickly.
Maximize your model velocity
Understand how well you are positioned to achieve model velocity and get suggested areas of improvement by taking this free 10-minute assessment
Our data scientists feel right at home on the Domino platform and appreciate how much freedom they have in experimenting in Domino.
Global Head of AI and Machine Learning
Accelerating Digital TransformationRead The Story
Frequently Asked Questions
What data science tools are supported in Domino's Workbench?
We are continuously introducing new tools and integrations that can be used in Domino. For a full list of tools currently supported in our tech ecosystem, click here.
How does Domino detect model drift?
Domino's model monitoring capabilities automatically detect and track data drift in the model’s input features and output predictions. Users can select among a variety of statistical checks to best suit their monitoring needs. If you have ground truth data for the model, Domino can ingest it to calculate and track the model’s prediction quality using standard measures such as accuracy, precision, and more.
What models can be monitored in Domino?
Models built and hosted in Domino as Model APIs are monitored automatically. Models built and hosted externally to Domino in popular languages, frameworks (e.g., Python, R, SAS, MATLAB, TensorFlow, DataRobot), and environments can be registered in Domino for monitoring.