About the role
- • Own and operate the on-premise Kubernetes clusters across PR, DR, and UAT — upgrades, patching, capacity, and incident response.
- • Build and maintain on-prem CI/CD pipelines and secure air-gapped deployment workflows.
- • Implement and maintain the Prometheus/Grafana monitoring stack and an on-prem logging solution, with alerting for critical workloads.
- • Work closely with application and data teams (streaming, database) to keep the platform secure, available, and performant.
- • Document infrastructure, runbooks, and troubleshooting procedures for a security-hardened, air-gapped setup.
1.On-Premise Kubernetes (End-to-End): • Proven hands-on experience installing, configuring, and operating a Kubernetes cluster entirely on-premise / bare-metal (e.g. RKE2, kubeadm, k3s) — not managed/cloud K8s (EKS/GKE/AKS).• Strong grasp of on-prem-specific components: CNI (e.g. Calico), MetalLB or equivalent for bare-metal load balancing, ingress controllers, and persistent storage (e.g. Longhorn, local-path-provisioner).• Experience managing multi-node clusters across Production, DR, and UAT with high-availability and DR failover considerations.• Comfortable troubleshooting at the cluster, node, and container-runtime level (e.g. containerd) in a fully air-gapped environment2. On-Premise DevSecOps Tooling [MANDATORY] : • Experience standing up and administering CI/CD tooling on-premise — Jenkins or equivalent (GitLab CI, Tekton, etc.), fully self-hosted with no external/SaaS dependency.• Ability to design secure, air-gapped CI/CD and artifact/image workflows (e.g. offline image transfer and import pipelines) where the environment has no direct internet access.• Working knowledge of security hardening for on-prem environments — access controls, sudo/user policy restrictions, firewall and network segmentation across subnets.
3. On-Premise Monitoring & Logging [MANDATORY]• Hands-on experience deploying and managing a self-hosted monitoring and alerting stack — Prometheus and Grafana at minimum.• Experience with an on-prem logging/log-aggregation stack (e.g. Loki, ELK/EFK, or equivalent) for cluster and application-level observability.• Ability to build actionable dashboards and alerts for cluster health, node resource usage, and workload-level metrics without relying on cloud-native/SaaS monitoring tools.
4. Linux Fundamentals [MANDATORY]• Strong Linux administration skills (RHEL/CentOS preferred) — Kubernetes runs on Linux, and day-to-day troubleshooting happens at the OS level.• Comfortable with systemd, networking (iptables/nftables, routing), storage/disk management, and package management in an offline/air-gapped context.• Solid shell scripting ability for automation and troubleshooting.