Senior GPU-Accelerated Kubernetes Runtime Engineer

2100 NVIDIA USA; United States; Remote; USD 184,000 - 350,000; Full time

Job summary

NVIDIA Corporation is seeking a Software Engineer for the Accelerated Kubernetes Runtime team to design and build automation systems powering GPU-accelerated distribution across NVIDIA AI accelerators.

You will develop Kubernetes controllers, CRDs, and operators to automate installation, upgrades, and rollbacks across thousands of clusters, delivering reliable, scalable, self-service tooling for researchers and engineers.

Qualifications

• Bachelor's in Computer Science or equivalent; 8+ years professional experience with at least 3 years on Kubernetes development.
• Experience building production Kubernetes systems with controllers, operators, and CRDs.
• Strong Go proficiency and building scalable Go services for distributed systems.
• Hands-on Helm, Kustomize, and managing Kubernetes manifest packaging and templating.
• Demonstrated ability to design automation systems that replace manual processes with self-service tooling.

Responsibilities

• Design and implement runtime features that orchestrate the lifecycle of runtime components across thousands of Kubernetes clusters without manual intervention.
• Build and maintain systems that configure, package, validate, and distribute accelerated compute components.
• Develop Kubernetes controllers, CRDs, and operators that automate runtime installation, upgrade, and rollback operations with API driven workflows.

Skills

Kubernetes development

Go programming

CRDs / CustomResourceDefinitions

Controllers / Operators

Distributed systems design

Automation / self-service tooling

API design

Helm

Kustomize

Education

Bachelor's degree in Computer Science or equivalent

Tools

Go

Helm

Kustomize

Job description

NVIDIA Corporation is seeking a Software Engineer for the Accelerated Kubernetes Runtime team to design and build automation systems powering GPU-accelerated distribution across NVIDIA AI accelerators.

You will develop Kubernetes controllers, CRDs, and operators to automate installation, upgrades, and rollbacks across thousands of clusters, delivering reliable, scalable, self-service tooling for researchers and engineers.