Productionizing AI in Enterprises: MLOps 6.2

[lhaviland]

(Part of an MLOps learning session series.)

DataRobot MLOps focuses on allowing your organization to adopt AI at scale, on your terms. It enables production model deployment and monitoring, model lifecycle management, and robust model governance. In addition to lifecycle management for DataRobot models, DataRobot MLOps works with your custom models (e.g., R, Python, Java, Keras) and within your own infrastructure (AWS, GCP, Azure, etc.).

In this learning session, we review DataRobot MLOps with an emphasis on the valuable new features in the 6.2 release. For example, you can now containerize your DataRobot models and get predictions on your own infrastructure. Also, model monitoring now offers segmented analysis. Finally, in this release we added new governance workflows for managing all your models. Join us to learn about these features and more and ask questions of our hosts.

Hosts

• Tristan Spaulding (DataRobot, Product Manger)
• Chris Cozzi (DataRobot, Product Manager)
• Rajiv Shah (DataRobot, Data Scientist)
• Jack Jablonski (DataRobot, AI Success Manager)

More Information

See the 6.2 MLOps demos in the What's New in Release 6.2 announcement:

• Segmented Analysis
• Usability Enhancements
• Portable Prediction Servers (PUBLIC BETA for 6.2)
• Custom Model Dependencies 
• Custom Model Testing Enhancements

Also, have a look at the Machine Learning Operations (MLOps) Walkthrough

Let us know what you think!

Have questions not answered during the learning session? Want to continue your conversation with Tristan, Chris, and Rajiv? You can send email to learning_sessions@datarobot.com or Post Your Comment here. We're looking forward to hearing from you!

Learning Session- MLOps 6.2 What's New.pdf ‏1 KB