Supercomputing, Software Engineer - Scheduling

at  OpenAI

ABOUT THE TEAM

The Supercomputing Scheduling Pillar at OpenAI is dedicated to ensuring the reliability, scalability, and user-friendliness of job lifecycle management, with an emphasis on efficient and flexible job scheduling, quota management, and job execution workflows. We maximize researcher productivity by ensuring high goodput, efficient packing, and a consistent, ergonomic training workflow, while scaling to ever larger supercomputers while reducing operational burden to the team.

ABOUT THE ROLE

As an engineer in the Scheduling Pillar, you will design, write, deploy, and operate job lifecycle management systems for model training on some of the largest supercomputers in the world. The scale is immense, the timelines are tight, and the organization is moving fast; this is an opportunity to shape a critical system in support of OpenAI’s mission.
This role is based in San Francisco, CA. We use a hybrid work model of 3 days in the office per week and offer relocation assistance to new employees.

IN THIS ROLE, YOU WILL:

• Design, implement and operate components of our quota management, job scheduling, and queuing systems. In short, you will work on the key user interface to our supercomputers.
• Interface with researchers to understand workload requirements
• Harmonize job lifecycle features with cluster infrastructure, storage, and hardware health requirements.

YOU MIGHT THRIVE IN THIS ROLE IF YOU:

• Have significant experience with hyperscale scheduling systems
• Possess strong programming skills
• Have experience working in public clouds (especially Azure)
• Execution focused mentality paired with a rigorous focus on user requirements
• As a bonus, have an understanding of AI/ML workloads