Modern apps. Maximum reliability.
What’s included:

• Cluster design (EKS/AKS/GKE/vanilla), node pools, autoscaler, CNI networking.

• Helm charts/operators, namespaces/quotas, policy-as-code (OPA/Gatekeeper).

• CI/CD pipelines, progressive delivery (Argo Rollouts/Flagger), secrets management.

• Observability: metrics/logs/traces (Prometheus/Grafana/Cloud-native), SLO/SLI.

• Security: image scanning (Trivy/Grype), PodSecurityStandards, minimal RBAC.

• Backups: etcd/persistent volumes, DR playbooks.

Deliverables:

• Cluster template (IaC), baseline set of charts/manifests.

• Deployment/upgrade guide + naming standards.

• SLO set and alert matrix per component.

Process: PoC → cluster build → baseline services → workload migration → stabilization.

Success metrics: release speed, MTTR, % automated deployments, resource utilization.

Options: service mesh (Istio/Linkerd), KEDA, Kyverno, multi-cluster/global ingress.