Architect Scalable Kubernetes Platforms with GitOps

This skill acts as an expert Kubernetes architect specializing in cloud-native infrastructure, advanced GitOps workflows (ArgoCD/Flux), and enterprise container orchestration at scale. It masters Kubernetes across the major managed providers (EKS, AKS, GKE) and on-premises deployments, and it aims at scalable, secure, and cost-effective platform engineering that improves developer productivity.

Its instructions run in four steps: gather workload requirements, compliance needs, and scale targets; define cluster topology, networking, and security boundaries; choose GitOps tooling and a delivery strategy for rollouts; and validate with staging before defining rollback and upgrade plans. Its safety guidance is explicit: avoid production changes without approvals and rollback plans, and test policy changes and admission controls in staging first.

Its capabilities cover twelve areas: managed and self-managed Kubernetes platforms (including kubeadm, kops, kubespray, and air-gapped deployments); GitOps and continuous deployment with ArgoCD, Flux v2, Jenkins X, and Tekton, plus progressive delivery through Argo Rollouts and Flagger; infrastructure as code with Helm 3.x, Kustomize, Jsonnet, cdk8s, and Terraform/OpenTofu, and policy as code with OPA, Gatekeeper, and Kyverno; cloud-native security spanning Pod Security Standards, network policies, runtime security tools like Falco, and supply chain practices such as SLSA and Sigstore; service mesh architecture with Istio, Linkerd, Cilium, Consul Connect, and the Gateway API; container and image management including registries, multi-stage builds, and Tekton or Kaniko build strategies; observability with Prometheus, Thanos, Loki, Jaeger, and Grafana; multi-tenancy and platform engineering, including namespace strategies, RBAC design, and operator development with CRDs; scalability features such as HPA, VPA, Cluster Autoscaler, and KEDA; cost optimization with tools like KubeCost and OpenCost; and disaster recovery with Velero backups, multi-region deployment, and chaos engineering.

It grounds GitOps in the CNCF OpenGitOps principles: declarative, versioned and immutable, pulled automatically, and continuously reconciled. Its nine-step response approach moves from assessing workload requirements through designing the architecture, implementing GitOps, configuring security, setting up observability, planning scalability and multi-tenancy, optimizing cost, and documenting the platform.

Use it for designing Kubernetes platform architecture or a multi-cluster strategy, implementing GitOps workflows and progressive delivery, planning service mesh, security, or multi-tenancy patterns, or improving reliability, cost, or developer experience in Kubernetes.

Do not use it when you only need a local dev cluster or single-node setup, when you are troubleshooting application code without platform changes, or when you are not using Kubernetes or container orchestration at all.

Inputs are workload requirements, compliance needs, and scale targets. Outputs are a cluster topology and networking design, a chosen GitOps toolchain and rollout strategy, security policy configuration, an observability stack, autoscaling and multi-tenancy setup, cost optimization changes, and platform documentation with rollback and upgrade plans.

It is for platform engineers and Kubernetes architects designing or scaling enterprise container platforms, particularly where GitOps, multi-cluster resilience, security by default, and cost efficiency all need to be designed together.