Here are the high-level tasks you need to perform to generate a starter pipeline configuration:

Create infrastructure resources – Your pipeline requires certain AWS resources, for example the IAM user and roles with necessary permissions, an Amazon S3 bucket, and optionally an Amazon ECR repository.

Connect your Git repository with your CI/CD system – Your CI/CD system needs to know which Git repository will trigger the pipeline to run. Note that this step may not be necessary, depending on which combination of Git repository and CI/CD system you are using.

Generate your pipeline configuration – This step generates a starter pipeline configuration that includes two deployment stages.

Commit your pipeline configuration to your Git repository – This step is necessary to ensure your CI/CD system is aware of your pipeline configuration, and will run when changes are committed.

After you've generated the starter pipeline configuration and committed it to your Git repository, whenever someone commits a code change to that repository your pipeline will be triggered to run automatically. 

https://docs.aws.amazon.com/serverless-application-model/latest/developerguide/serverless-generating-example-ci-cd.html