Automating with AWS Step Functions
AWS Step Functions let you define and run workflows based on state machines (https://aws.amazon.com/step-functions/). A state machine is a combination of steps, which can be sequential, parallel, or conditional. Each step receives an input from its predecessor, performs an operation, and passes the output to its successor. Step Functions are integrated with many AWS services, such as Lambda, DynamoDB, and SageMaker, and you can easily use them in your workflows.
State machines can be defined using JSON and the Amazon States Language, and you can visualize them in the service console. State machine execution is fully managed, so you don't need to provision any infrastructure to run.
When it comes to SageMaker, Step Functions has a dedicated Python SDK, named the Data Science SDK (https://github.com/aws/aws-step-functions-data-science-sdk-python). In my humble opinion, this is a confusing name, as the SDK has nothing to do with data science...
