Senior AI Infrastructure Engineer — Scale GPU Clusters Remote

AI Breaking Wire; San Francisco, Northern (CA, KY); Hybrid; USD 280,000 - 420,000; Full time

Benefits offered by this job

Equity options

Health, vision, dental benefits

Unlimited PTO

Remote work options

Learning and development stipend

Job summary

OpenAI is seeking a Senior Infrastructure Engineer to design and scale ultra-high-performance distributed systems for training and inference of massive neural networks. You will optimize GPU cluster utilization, tackle networking bottlenecks, and collaborate with research teams to streamline experiments and model deployment.

The role requires deep expertise in HPC and distributed systems, with proficiency in C++, Python, CUDA, and Kubernetes, plus experience managing large-scale GPU clusters.

Qualifications

• Requires a degree (B.S./M.S./Ph.D.) in Computer Science or related field with 5+ years of industry experience.
• Deep expertise in distributed systems, HPC, and networking (InfiniBand, RoCE).
• Proficiency in C++, Python, CUDA, and modern orchestration tools like Kubernetes.
• Experience managing large-scale cloud or on-prem GPU clusters.

Responsibilities

• Architect, build, and scale ultra-high-performance distributed systems for training and running massive neural networks.
• Optimize GPU cluster utilization, networking bottlenecks, and memory hierarchies for large-scale training runs.
• Collaborate with research teams to streamline experiment velocity and model deployment.
• Troubleshoot complex infrastructure issues across thousands of accelerators.

Skills

Distributed systems

HPC

Networking

C++

Python

CUDA

Kubernetes

Education

BS/MS/PhD in Computer Science

Tools

InfiniBand

RoCE

Job description

OpenAI is seeking a Senior Infrastructure Engineer to design and scale ultra-high-performance distributed systems for training and inference of massive neural networks. You will optimize GPU cluster utilization, tackle networking bottlenecks, and collaborate with research teams to streamline experiments and model deployment.

The role requires deep expertise in HPC and distributed systems, with proficiency in C++, Python, CUDA, and Kubernetes, plus experience managing large-scale GPU clusters.