Skip to main content

         This documentation site is for previous versions. Visit our new documentation site for current releases.      

Enrich your decision strategies with predictive models from Amazon SageMaker and (8.4)

Updated on May 3, 2021

As part of Pega’s Open AI initiative, users have always been able to import PMML-compliant models from third-party tools, to assist with their customer predictions.

In Pega Marketing 8.3 we doubled-down on this concept with a Real-Time AI Connector to Google Cloud Machine Learning. This is a set of APIs that allows clients to initiate service callouts to Google Cloud Machine Learning models (TensorFlow and XGBoost) in real-time, during their next-best-action decisions.

In 8.4, we’ve taken Open AI to the next level, adding a Real-Time AI Connector to Amazon SageMaker, as well as providing a direct integration with The integration with means that users can seamlessly import H2O-3 and H2O Driverless AI models directly into their Pega applications, and run them natively in their next-best-action strategies.

The following video illustrates how to import H2O models to Prediction Studio:

"Video showing how to import an H2O model to Prediction Studio"
Selecting and importing an H2O model to Prediction Studio

For more information, see Importing a predictive model.

The following video illustrates how to connect to Amazon SageMaker models in Prediction Studio:

"Video showing how to import an Amazon SageMaker model to Prediction Studio"
Connecting to an Amazon SageMaker model in Prediction Studio

For more information, see Connecting to an external model.

  • Previous topic Ease into AI-driven, event-triggered engagement with Multi-Schedule Outbound (8.4)
  • Next topic Implement agile change management with 1:1 Operations Manager (8.3)

Have a question? Get answers now.

Visit the Support Center to ask questions, engage in discussions, share ideas, and help others.

Did you find this content helpful?

Want to help us improve this content?

We'd prefer it if you saw us at our best. is not optimized for Internet Explorer. For the optimal experience, please use:

Close Deprecation Notice
Contact us