How to create multi-account ML workflows with Amazon SageMaker Pipelines (Level 300)
Machine learning workflows are hard to build because you need to create hundreds of code packages for data preparation, model training and deployment, and stitch them together so they run as a sequence of steps. In this session, learn about Amazon SageMaker Pipelines, the world’s first machine learning CI/CD service designed to be accessible for every developer and data scientist. Amazon SageMaker Pipelines brings CI/CD pipelines to machine learning, reducing the months of coding previously required to just a few hours. We show how you can create a multi-account deployment pipeline using a custom SageMaker project template and discover how Amazon SageMaker Pipelines manages dependencies and orchestrates the workflow.
Julian Bright, Senior AI/ML Specialist Solutions Architect, AWS