DataRobot Summer 2024

Scale and Maintain the Integrity of all your Predictive AI Models

New predictive AI functionality to help you scale predictive AI solutions and ensure model integrity and performance from build through deployment for DataRobot built models or custom models.

Create High-Quality Predictive Models Fast

Quickly prepare data for modeling and evaluate model performance when building.

• Secure Data Connectivity in NextGen: Find, share, and leverage data easily with enhanced browsing and preview functionality, profile details, in cloud data warehouses, cloud storage, and the AI Catalog.
• Wrangle, Join, and Aggregate in NextGen: Enhance your data workflows by seamlessly joining, aggregating, and transforming data directly on supported cloud data warehouses or data stored in the DataRobot AI Catalog and blob storage.
• Feature Discovery in NextGen: Build custom Feature Discovery recipes for your specific use cases to generate new datasets with derived features. These datasets can then be used to unlock deeper insights via exploration, wrangling, or experimentation. 
• Explain Predictions with SHAP Insights in NextGen: Quickly understand predictions with enhanced SHAP explanations support for all model types and new individual PE functionality that calculates SHAP values for each individual row.
• Slices Insights: Enhance your understanding of how models perform on different subpopulations by viewing and comparing insights based on segments of your project data. Slice data by date/time, numerical, categorical, and boolean data types. 
• Enhanced Confusion Matrix in NextGen: Train classifiers on datasets with unlimited classes within Workbench, then quickly understand the effectiveness of your classifiers with our enhanced confusion matrix.
• Side-by-Side Modeling Insights in NextGen: Rapidly improve model performance by easily assessing model performance and comparing models across experiments, even those that use varied datasets and modeling parameters.
• Time Series Experience in NextGen: Easily build robust, fine-grained time series forecasts in Workbench.
• ​​Codespaces and Codespace Scheduling in NextGen: Build reusable automated workflows with new Codespace features. Open, view, edit, and execute multiple notebook and non-notebook files in the same container session. Easily establish automated jobs at any desired cadence. Monitor your scheduled notebook jobs and track run history. Configure scheduled notebooks to develop automated, reusable workflows for effortless execution.
• Project Migration from Classic to NextGen: Easily move projects and datasets from DataRobot Classic into NextGen to take full advantage of all of the new functionality, visuals, and collaboration features of NextGen.
• Scale Enhancements: Seamless handling of large datasets throughout the ML lifecycle with incremental learning and enhanced NVIDIA GPU compatibility.

Organize, Track and Manage all Custom Predictive Models with Ease

Standardize how you organize and manage all your custom models for unified governance, oversight, and control.

• Track and Organize All Custom Models: Standardized testing, version control, and compliance docs for every custom ML model, from custom time series models to computer vision and NLP models to geospatial. 
• MLFlow Integrations: Accelerate AI experimentation by importing all charts, metadata, and parameters directly from MLFlow.  
• Global Models for Predictive Use Cases: Increase your impact and standardize performance by creating custom global models then share the best performing models with everyone in your organization. 
• Feature Impact for Custom Metrics: Clearly see which features are driving model decisions and easily identify opportunities to improve model performance in production. 
• Custom Jobs: Create the experience that your team and your projects need by customizing all aspects of your workflows, including validationing, model performance, metrics, notification jobs, retraining, and data pipelines.

Maintain Predictive AI Model Integrity in Production

Deliver and maintain meaningful models faster and with fewer headaches with unified production workflows and monitoring customized to your goals.

• Serverless NextGen Predictions: Enjoy real-time and batch predictions without needing extra server resources with Serverless NextGen Predictions. Handle large data sizes and run more jobs simultaneously, whether ad-hoc or scheduled, on the same cluster footprint. 
• Batch Tracking for Monitoring: Track changes in KPIs directly to their cause with our new Accuracy Batch Selection interface. Effortlessly monitor and analyze batch accuracy, visualize performance trends and effectively manage KPIs over time. 
• Hosted Custom Predictive Metrics: Create custom accuracy, drift, cost, readability, and other metrics. 
• Notification Policies: Manage your team’s time and focus more effectively with customized notification policies. Define recipients of notifications, notification channels, and frequency of notifications, plus receive real-time notifications. Save even more time with notification policy templates that can be reused across deployments.
• Custom Retraining: Create your perfect model with fine-tuned retraining. Use your own code in any language to tailor your model training process with custom policies from build through management. Whether you’re using experiments or custom models, you can leverage parameters, network access, and key values to deliver even more value.
• Enhanced Production Pipelines and Ecosystem Integrations: Unify production workflows with expanded integrations. Effortlessly build and manage production-grade deployment and monitoring pipelines with Airflow. Benefit from centralized management and monitoring while deploying to your preferred cloud platform. Enjoy seamless integration with Amazon Sagemaker and Azure Machine Learning for enhanced performance and compatibility.

Go the DataRobot Documentation Release Center for more information.