Software Engineer, Frontier Clusters Infrastructure

About this role

OpenAI seeks a Software Engineer to build and operate world-class supercomputer clusters supporting frontier AI model training. This role combines distributed systems expertise with hands-on infrastructure work, requiring you to scale Kubernetes at massive scale, automate bare-metal deployments, and ensure reliability across multi-datacenter environments.

What you'll do

• Design and scale large Kubernetes clusters with automation for provisioning, bootstrapping, and lifecycle management
• Build software abstractions that unify multiple clusters and provide seamless interfaces for training workloads
• Manage bare-metal node bring-up including firmware upgrades and repeatable deployment at scale
• Optimize operational metrics like reducing restart times and accelerating upgrade cycles
• Integrate networking and hardware health systems for end-to-end infrastructure reliability
• Develop monitoring and observability systems to detect issues early and maintain stability under extreme load

What they're looking for

• Kubernetes cluster operations and scaling in hyperscale environments
• Python, Go, or similar programming languages
• Infrastructure-as-Code tools (Terraform, CloudFormation)
• Bare-metal Linux and GPU hardware management
• Large-scale networking and data center operations
• Distributed systems engineering
• Cloud infrastructure (compute, networking, storage, security)
• GPU workloads and high-performance computing (bonus)