The GitLab Runner Infrastructure Toolkit (GRIT) is to be the primary entry-point for setting up and managing runner deployments. There are many, many ways to setup runner and it can be confusing. Several solutions have been developed in parallel and have overlapping scope. For example GitLab.com has a set of Terraform templates to manage their SaaS runners. But they are not publicly consumable. There is an official GitLab Helm chart as well as a Kubernetes Operator. And the docs have instructions on setting up runner, but no prescriptive starting point. GRIT intents to be that prescriptive starting point. Begin with GRIT and it will point you in the right direction, whether your setting up Linux, MacOS or Windows runners. Whether you're setting a development environment, production or end-to-end tests. Whether you're using autoscaling, using VMs or Kubernetes, using a HA setup with blue-green deployments. It will provide primarily Terraform templates that can be imported and parameterized as a library. And some scripting (as little as possible) to fill in the gaps. It will use the official Helm chart and K8s operator for setting up Kubernetes. But will also be able to setup the Fleeting stack inside or outside K8s. The overall goal is reduce repetition and increase dogfooding. We want our customers to deploy runner in exactly the same way we do on GitLab.com so they will benefit from our experience. And we will feel their pain. https://gitlab.com/gitlab-org/ci-cd/runner-tools/grit