Running machine learning projects in production requires a lot of orchestration. A popular way to schedule this is with Airflow, however it's not really optimized for ML workloads. The new kid on the block is Kubeflow Pipelines (part of Kubeflow). From their website: Kubeflow Pipelines is a platform for building and deploying portable, scalable machine learning (ML) workflows based on Docker containers.

We will start the workshop by providing context and an understanding of what Kubeflow Pipelines is, and when to use it. After this, there will be an intensive hands-on part in which you will create and deploy your own ML pipelines.