Feast provides a simple framework for defining and serving machine learning features. In order to serve features reliably, with low latency and at high scale, Feast relies heavily on cloud infrastructure such as DynamoDB or AWS Lambda. This talk explains how Feast manages its serving infrastructure reliably and predictably for users. Similar to Terraform, Feast exposes commands that allow users to understand and precisely control the set of infrastructure that will be deployed to serve their features. This talk will also discuss how this process of managing data infrastructure with Feast can fit into a CICD pipeline.