Wrap Up: Projects in Production


You’ve just completed the Projects in Production course, where you gained experience creating project bundles on the Design node and deploying them to the Automation node via the Project Deployer.

Here are a few of the main takeaways from this course:

  • In order to put a batch-processing project into production, you first create a project bundle on the Design node, and then deploy it to the Automation node via the Project Deployer.

  • Once you’ve mastered the fundamentals of pushing project bundles from the Design to the Automation node, you can use all of the same automation tools (scenarios with metrics, checks, reporters, and triggers) for MLOps purposes.

  • For example, you can create scenarios that send alerts about model performance metrics as logged in a model evaluation store, and package them in a bundle so the scenario can run in a production environment.

  • As your MLOps setup becomes more sophisticated, you can use scenarios to automatically create new project bundles and update deployments—provided you have the necessary degree of metrics and checks in place to have confidence in this process.

Learn More

Now that you have completed this course, you may wish to consult the product documentation for more information on production deployment and bundles.

When ready, continue progressing on your MLOps Practitioner learning path.