Site Reliability Engineer, Frontier Systems Infrastructure

About this role

OpenAI seeks a Site Reliability Engineer to build and operate hyperscale supercomputer clusters that power frontier AI model training. You'll manage massive Kubernetes deployments, automate bare-metal infrastructure, and ensure reliability across distributed data centers.

What you'll do

• Scale and provision large Kubernetes clusters with automation for lifecycle management
• Develop software abstractions that unify multiple clusters for training workloads
• Manage bare-metal node bring-up, firmware upgrades, and deployment automation
• Optimize operational metrics like cluster restart times and upgrade cycles
• Build monitoring and observability systems for early issue detection
• Integrate networking and hardware health systems for end-to-end infrastructure reliability

What they're looking for

• Kubernetes cluster operations and scaling at hyperscale
• Python, Go, or similar programming/scripting languages
• Infrastructure-as-Code tools (Terraform, CloudFormation)
• Linux, bare-metal systems, and GPU hardware
• Large-scale networking and cloud infrastructure
• Distributed systems engineering
• Automation and DevOps practices
• High-availability system design