Wrap Up: Preparing for Production

Congratulations!

You’ve just completed the Preparing for Production course, where you reviewed best practices for automation, pipeline optimization, and project documentation.

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

• Before any Dataiku project can begin its journey into production, a robust set of metrics, checks, data quality rules, scenarios, triggers, and reporters should be established in the project on the Design node.

• Before moving a Dataiku project into production, you should consider several refactoring and optimization techniques. These include:

  • removing objects no longer required;

  • adding Flow zones for greater readability;

  • introducing variables for better maintainability;

  • reviewing the project’s efficiency—namely, the optimal use of computational resources and recipe engines;

  • and using partitioning if the use case is appropriate.

• Good documentation is an important component for a project to be successful in production. A well-documented workflow aids reproducibility, is easier to maintain, and allows better collaboration between team members. One major tool to support this goal is a wiki.

Learn more

Now that you have completed this course, you may wish to consult the reference documentation for more information on topics like automation, partitioning, or wikis.

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