Workload Porting & Performance Engineer

📋 Description

• • The Workload Porting & Performance Engineer role at OpenAI is critical for evaluating new hardware platforms by porting benchmarks and real-world AI workloads, analyzing performance, and identifying system bottlenecks to ensure platforms meet performance expectations.
• • Day to day, the engineer will port and enable workloads on new systems, characterize performance across compute, memory, storage, and networking, develop and run performance experiments, profile and debug issues across the software-hardware stack, and provide actionable feedback to hardware architecture, performance modeling, and software engineering teams.
• • The Infrastructure team at OpenAI builds and evaluates systems that power advanced AI workloads, collaborating closely with hardware, modeling, and architecture teams to bridge the gap between theoretical capability and observed system performance on evolving platforms.
• • In this role, the engineer will deepen expertise in performance analysis, workload optimization, and system-level debugging, gain hands-on experience with cutting-edge AI hardware, and contribute directly to platform readiness and deployment decisions for next-generation AI infrastructure.