Platform Engineer, On-Prem Kubernetes & VM Fabric

Referment; Greater London; On-site; GBP 90,000 - 120,000; Full time

Job summary

Referment is building an on‑prem Kubernetes platform anchored on a Linux KVM/libvirt VM fabric. You will design, implement and operate from hypervisor through to cluster lifecycle, ensuring security, performance and reliability in a self‑managed environment.

You will own networking, storage, CSI integration, and multi‑tenancy patterns while enabling GitOps workflows with ArgoCD/Flux and CIS‑level hardening. This role suits an engineer who thrives in deep, end‑to‑end system ownership.

Qualifications

• : Solid production experience running Kubernetes in a self‑managed on‑premises context, understanding the risks without a managed control plane.
• : Hands‑on experience designing and operating a Linux KVM/libvirt‑based VM fabric for Kubernetes, ideally from an early stage.
• : Strong Linux systems administration background: networking, storage, service management, kernel tuning.

Responsibilities

• Design, build and operate on‑prem Kubernetes platform across VM fabric and cluster lifecycle.
• Own hypervisor to Kubernetes integration, including networking, storage and CNI handoff.
• Manage etcd, backups, disaster recovery, upgrades with minimal workload disruption.

Skills

Kubernetes

Linux admin

KVM/libvirt

Networking

Storage

Terraform

Ansible

Packer

GitOps

CI/CD

Prometheus/Grafana

RBAC/Security

Troubleshooting

CI/CD pipelines

Ceph/Longhorn

Etcd

Tools

kubeadm

Cluster API

Kubespray

Terraform

Ansible

Packer

Job description

Referment is building an on‑prem Kubernetes platform anchored on a Linux KVM/libvirt VM fabric. You will design, implement and operate from hypervisor through to cluster lifecycle, ensuring security, performance and reliability in a self‑managed environment.

You will own networking, storage, CSI integration, and multi‑tenancy patterns while enabling GitOps workflows with ArgoCD/Flux and CIS‑level hardening. This role suits an engineer who thrives in deep, end‑to‑end system ownership.