Kubernetes platform engineer

• Build, administer, and troubleshoot Kubernetes platforms supporting AI and data-intensive workloads.
• Diagnose failures across control plane components, kubelet, CNI, CSI, ingress, service discovery, scheduling, node lifecycle, container runtime, and resource isolation.
• Support GPU-enabled Kubernetes environments, including device plugin behavior, driver dependencies, node health, and workload placement.
• Improve reliability through automation, standardized configuration, upgrade planning, and cluster validation gates.
• Investigate issues involving storage throughput, network policy, DNS, image pulls, autoscaling, pod eviction, and degraded node states.
• Partner with Linux, networking, validation, and SRE teams to resolve cross-layer failures.
• Create reusable runbooks, dashboards, and health checks for day-2 support.
• Contribute to platform hardening, tenant readiness, and service-level objectives.

• 7+ years of infrastructure engineering experience.
• Deep hands-on Kubernetes administration experience.
• Strong understanding of Kubernetes internals and cluster troubleshooting.
• Experience with container runtimes, Helm, GitOps or declarative operations, and cluster lifecycle management.
• Experience supporting GPU workloads on Kubernetes.
• Strong Linux administration background.
• Understanding of data-center networking dependencies.
• Ability to debug from symptom to root cause across node, pod, network, storage, and control-plane layers.
• Scripting and automation skills in Python, Bash, or Go.

Preferred

• Kubeflow
• Argo
• Prometheus
• Grafana
• Loki
• Service mesh technologies
• Bare-metal Kubernetes
• High-performance storage integration
• Regulated or high-change-control production environments