Local development can be slow and debugging difficult when training machine learning models. This hands-on lab will demonstrate how to your development experience by enabling you to seamlessly run your training script locally or scaled out on the cloud without any code changes. Attendees will learn best practices for structuring code to isolate the train ing script, leveraging containers for environment consistency, and integrating with Amazon SageMaker.