CI/CD Pipelines for Kubernetes

Most CI/CD pipelines live outside the cluster they deploy to. A GitHub Actions runner builds your image, pushes it to a registry, and shells out to kubectl apply. It works, but your pipeline exists in a completely different world from your workloads. Different auth model, different scaling model, different observability. Kubernetes-native CI/CD tools close that gap by defining pipelines as Custom Resource Definitions that run inside the cluster, using the same Pods, volumes, and RBAC you already manage.

The KCNA exam's Cloud Native Application Delivery domain (16% weight) expects you to understand CI/CD concepts and how they fit into the Kubernetes ecosystem. This article covers the CI side: how pipelines build and test container images, how Tekton and Argo Workflows model that process as Kubernetes resources, and where external tools like GitHub Actions fit in. The next article covers the CD side with GitOps.