Senior Kubernetes Engineer for GPU AI Compute Platform

GTN Technical Staffing; Dallas (TX); On-site; USD 140,000 - 200,000; Full time

Job summary

GTN Technical Staffing is seeking a Senior Kubernetes Engineer to design and scale GPU-accelerated compute platforms across on-prem and hybrid environments. You will own the architecture and operation of large Kubernetes clusters, enabling high-throughput, multi-tenant workloads with automation and platform evolution.

You will work with platform, HPC, and ML teams to deploy GPU-enabled clusters, extend Kubernetes via operators, and optimize NVIDIA tooling like GPU Operator, DCGM, and device

Qualifications

• Strong experience operating Kubernetes in large-scale, production environments.
• Hands-on experience with NVIDIA GPU ecosystem (GPU Operator, device plugins, MIG, DCGM).
• Proficiency in Go or Python for Kubernetes operators and automation.
• Deep understanding of Kubernetes internals: CRDs, controllers, RBAC, scheduling.

Responsibilities

• Design, deploy, and operate large-scale Kubernetes clusters optimized for GPU workloads.
• Architect container platforms for AI/ML, LLM training, and HPC use cases.
• Extend Kubernetes with custom operators and controllers for automation.
• Integrate NVIDIA ecosystem components and GPU scheduling strategies.
• Improve cluster efficiency with scheduler extensions and observability.

Skills

Kubernetes

GPU ecosystem

Go/Python

CRDs & controllers

GitOps CI/CD

Terraform/Helm/Kustomize

RBAC & multi-tenancy

Monitoring/Observability

Networking/CNI

Tools

ArgoCD

FluxCD

Terraform

Helm

Kustomize

Prometheus

Grafana

DCGM

Job description

GTN Technical Staffing is seeking a Senior Kubernetes Engineer to design and scale GPU-accelerated compute platforms across on-prem and hybrid environments. You will own the architecture and operation of large Kubernetes clusters, enabling high-throughput, multi-tenant workloads with automation and platform evolution.

You will work with platform, HPC, and ML teams to deploy GPU-enabled clusters, extend Kubernetes via operators, and optimize NVIDIA tooling like GPU Operator, DCGM, and device