GPU Compute Infrastructure Engineer

Linuxcareers; San Francisco, Northern (CA, KY); Hybrid; USD 210,000 - 270,000; Full time

Job summary

OpenAI seeks engineers to design and operate Kubernetes-based controllers that coordinate GPU compute infrastructure across sites. You will define APIs, build provisioning services, and manage lifecycle operations from boot to decommissioning, ensuring reliability and scalability.

Ideal candidates have strong distributed-systems background, experience with Kubernetes reconciliation, and the ability to troubleshoot complex cross-boundary issues with staged rollouts across nodes and racks.

Qualifications

• Strong software engineering fundamentals.
• Experience designing production distributed systems or infrastructure services.
• Experience with Kubernetes-based APIs and reconciliation patterns.
• Ability to design reliable APIs and asynchronous workflows.
• Capability to diagnose reliability and performance issues across borders.

Responsibilities

• Design, build, and operate Kubernetes-based controllers and services across sites.
• Define APIs and resource models for lifecycle operations across hardware providers.
• Build provisioning/configuration services for network boot, firmware, and host config.
• Develop lifecycle management for discovery, provisioning, upgrades, and recovery.
• Design recovery mechanisms for concurrent changes and partial failures with staged rollouts.

Skills

Distributed systems

Kubernetes

APIs design

Troubleshooting

Job description

OpenAI seeks engineers to design and operate Kubernetes-based controllers that coordinate GPU compute infrastructure across sites. You will define APIs, build provisioning services, and manage lifecycle operations from boot to decommissioning, ensuring reliability and scalability.

Ideal candidates have strong distributed-systems background, experience with Kubernetes reconciliation, and the ability to troubleshoot complex cross-boundary issues with staged rollouts across nodes and racks.