This document discusses using Puppet and AWS together to dynamically scale infrastructure. It provides examples of building out the necessary components like OS, software stack, networking, and application deployment in an automated and orchestrated way. Key steps include using CloudFormation to build out the VPC and networking components, baking AMIs with foundational configurations, deploying applications via Puppet modules, and using tools like AWS CodeDeploy and auto scaling for automated deployments and scaling. The overall goal is to enable continuous delivery of applications at high velocity by making infrastructure dynamic and mutable.