#go#github_actions#kubernetes#operator
Actions Runner Controller (ARC) is a tool that helps you automatically manage and scale self-hosted GitHub Actions runners using Kubernetes. It creates runner scale sets that grow or shrink based on how many workflows you are running, making your CI/CD process more efficient and cost-effective. ARC uses containers for runners, so new instances can start or stop quickly and cleanly. You can install ARC easily with Helm on Kubernetes and customize runners with features like custom images, volumes, and scripts. This automation saves you time and resources by matching runner capacity to your actual workload needs[1][2][3].
https://github.com/actions/actions-runner-controller
https://github.com/safarijv/kubelib
If you're adopting Kubernetes as an orchestration system for #machine_learning jobs, the last thing you want is for the mere act of using Kubernetes to create more problems than it solves. Kubelib provides a set of Pythonic interfaces to #Kubernetes, originally to aid with Jenkins scripting. But it can be used without Jenkins as well, and it can do everything exposed through the kubectl #CLI or the Kubernetes #API.