Inference Runtime, Engineering Manager

📋 Description

• Lead engineers in distributed systems and model architecture.
• Collaborate with ML researchers, engineers, and PMs to deploy tech.
• Contribute across infra to performance tuning.
• Introduce techniques and tools to improve inference performance.
• Build tools to identify bottlenecks and address high-priority issues.
• Optimize code and GPU fleet to maximize FLOPs and RAM.

🎯 Requirements

• Understanding of modern ML architectures and inference optimization.
• Own problems end-to-end; self-directed to fill gaps.
• 15+ years of software engineering experience.
• Familiar with PyTorch, NVIDIA GPUs, CUDA, NCCL; HPC tech (InfiniBand, MPI, NVLink).
• Experience architecting, building, observing, debugging production distributed systems.
• Refactor production systems at rapidly increasing scale.

🎁 Benefits

• Equal opportunity employer.
• OpenAI Affirmative Action and Equal Employment Policy.
• Reasonable accommodations for applicants with disabilities.
• OpenAI Global Applicant Privacy Policy.
• Background checks as required by law.