Edge AI Infrastructure Engineer – Kubernetes & GPU

UMATR; Austin (TX); On-site; USD 120,000 - 190,000; Full time

Job summary

UMATR is hiring an Infrastructure Engineer to own platform deployment and scaling across Kubernetes, GPU infrastructure and cloud environments. As an early engineer, you’ll shape architecture and tooling with high ownership from day one.

You’ll manage end-to-end deployment, build Kubernetes on-prem deployments, implement GitOps, and own CI/CD pipelines and AI inference workloads on customer hardware.

Qualifications

• Strong Kubernetes expertise with troubleshooting production environments.
• Hands-on with Kubernetes on-prem deployments (k3s, RKE2, MicroK8s or similar).
• Production experience with Terraform and Docker in CI/CD.
• Experience running AI/ML workloads on GPU infrastructure.
• Own CI/CD pipelines and software delivery improvements.
• Strong Linux/Bash skills and Python for applications.

Responsibilities

• Own end-to-end deployment lifecycle for customer environments from bare hardware to running systems.
• Build and maintain Kubernetes deployments across on-premise environments.
• Develop GitOps workflows and automated reconciliation.
• Create IaC-based cloud infrastructure for dev and staging.
• Own CI/CD pipelines, builds, tests, deployments and releases.
• Deploy and optimise AI inference workloads on GPU hardware.
• Tune model serving infrastructure for performance and latency.
• Improve observability via monitoring, tracing and diagnostics.
• Secure systems for secrets, permissions and industrial networks.
• Produce documentation, runbooks and automated tooling.
• Identify opportunities to reduce manual work and improve efficiency.

Skills

Kubernetes

CI/CD

Linux

Python

GPU workloads

Observability

GitOps

Tools

Terraform

Docker

Flux/Argo CD

OpenTelemetry

Job description

UMATR is hiring an Infrastructure Engineer to own platform deployment and scaling across Kubernetes, GPU infrastructure and cloud environments. As an early engineer, you’ll shape architecture and tooling with high ownership from day one.

You’ll manage end-to-end deployment, build Kubernetes on-prem deployments, implement GitOps, and own CI/CD pipelines and AI inference workloads on customer hardware.