Senior Kubernetes Platform Engineer

About the role

Firmus is looking for a Senior Kubernetes Platform Engineer to own the lifecycle and reliability of its multi‑tenant Kubernetes estate that powers AI workloads across all sites. The role is hands‑on, focused on automation, fleet‑scale operations and continuous improvement of the platform.

Key responsibilities

• Operate and continuously improve the multi‑tenant Kubernetes platform across all Firmus sites, ensuring high availability and compliance.
• Build and maintain automation tooling, guarded remediation and orchestration for fleet‑scale operations.
• Execute cluster lifecycle tasks – provisioning, patching, upgrading and decommissioning – using staged or canary release paths.
• Recover and troubleshoot Kubernetes control planes, including etcd state, certificate rotation and failed upgrades.
• Manage the virtual‑cluster isolation layer and automated tenant onboarding pipelines.
• Operate the GPU integration layer (e.g., NVIDIA GPU Operator), handling device plugins, scheduling and driver coordination.
• Diagnose scheduling, CNI, CSI and admission failures, driving permanent fixes through AI Infrastructure.
• Lead technical recovery during major incidents and mentor engineers on fault diagnosis and run‑book creation.

Required profile

• 8+ years of platform/infrastructure engineering experience with strong ownership of production Kubernetes platforms in a 24/7 environment.
• Deep knowledge of Kubernetes control‑plane internals, multi‑cluster management and fleet‑scale operations.
• Proven ability to act as a senior escalation point, including on‑call duties and incident leadership.
• Solid understanding of Kubernetes security fundamentals such as admission control, workload identity and network policy.

Required skills

• Kubernetes (control plane, etcd, API server, controllers, schedulers, certificate rotation)
• Writing Kubernetes controllers, operators or admission logic
• Multi‑tenant/virtual‑cluster patterns (e.g., vCluster)
• GPU‑enabled Kubernetes (NVIDIA GPU Operator, device plugins, GPU scheduling)
• Infrastructure‑as‑code and GitOps (OpenTofu, Terraform, Ansible, Argo CD)
• Scripting/programming for automation (Go, Python, Bash)