Kale enables data scientists to orchestrate end-to-end machine learning (ML) workflows.

Kale simplifies the use of Kubeflow, giving data scientists the tool they need to orchestrate end-to-end ML workflows. Kale provides a UI in the form of a JupyterLab extension. You can annotate cells in Jupyter Notebooks to define: pipeline steps, hyperparameter tuning, GPU usage, and metrics tracking. Click a button to create pipeline components and KFP DSL, resolve dependencies, inject data objects into each step, deploy the data science pipeline, and serve the best model.

See From Notebook to Kubeflow Pipelines to KFServing for a tutorial overview of Kale.