Software Engineer, Compute Foundations Systems

OpenAI

Where
San Francisco

Posted 4 months ago· Checked live at OpenAI 1 hour ago

Stale: open 121 daysUp for more than two weeks. It may already be filled.

See how you score for this job

One sentence and your resume. We score it against what OpenAI asks for, line by line.

See my score
Is this job still open?
Yes. We checked it live on OpenAI's own hiring page 1 hour ago.
What is the job?
Frontier Systems Foundations, part of Compute Foundations at OpenAI, builds the systems software foundation that turns new compute infrastructure into reliable, usable capacity for frontier model training. Our mission is to make some of the world's largest GPU clusters work reliably for frontier training.
What does it pay?
The posting doesn't say.
Our verdict
Stale. Up for more than two weeks. It may already be filled.

What the job is

Frontier Systems Foundations, part of Compute Foundations at OpenAI, builds the systems software foundation that turns new compute infrastructure into reliable, usable capacity for frontier model training.

Our mission is to make some of the world's largest GPU clusters work reliably for frontier training.

In this role, you will

  • Build and maintain the Linux host software stack for large GPU clusters, including Ubuntu and OS images, kernel configuration and modules, drivers, packages, disks and storage configuration, and machine configuration.
  • Design reproducible OS-image builds, package and repository workflows, and system configuration for heterogeneous bare-metal and cloud compute fleets.
  • Integrate, test, and qualify kernels, modules, drivers, packages, and firmware across new and existing hardware platforms; build safe canary, rollback, and recovery paths for system changes.
  • Bring up new hardware platforms and compute SKUs, working with hardware engineers and vendors to resolve firmware, driver, operating-system, and compatibility issues.
  • Debug complex system failures across boot and provisioning, firmware, disks, kernels and drivers, and workload interactions; turn recurring failure modes into durable fixes, tests, and automation.
  • Improve provisioning, repave, maintenance, and recovery correctness by eliminating manual host-by-host intervention and making system behavior predictable at fleet scale.
  • Build focused systems tooling and diagnostics that make the host software stack easier to validate, troubleshoot, and operate across the installed fleet.

You might thrive in this role if you

  • Have significant experience building, integrating, or operating production Linux systems, particularly Ubuntu, Debian, or another major Linux distribution.
  • Have deep experience in one or more of: Linux kernels, modules, or device drivers; Linux distribution, package, repository, or OS-image engineering; boot, provisioning, disks, or machine configuration; firmware and driver integration, qualification, or rollout.
  • Can write, debug, and maintain production-quality systems software and automation using appropriate systems languages, scripting, and Linux tooling.
  • Understand how to build, test, package, qualify, and safely deliver system-level changes, including compatibility testing, staged rollout, rollback, and recovery.
  • Can systematically debug failures that cross hardware, firmware, boot, operating-system, kernel, and driver boundaries, and collaborate effectively with hardware and software partners.
  • Enjoy investigating difficult system behavior, identifying underlying failure modes, and building pragmatic fixes that make production systems more reliable.

Bonus points if you

  • Have built bare-metal provisioning, machine-bootstrap, OS-reprovisioning, or system-lifecycle tooling.
  • Have worked with GPU systems, AI infrastructure, HPC clusters, accelerators, or other heterogeneous and performance-sensitive compute environments.
  • Have experience bringing up new server platforms or working directly with hardware and software vendors on firmware, driver, or operating-system compatibility issues.
  • Have contributed to Linux, kernel subsystems or modules, distributions, package ecosystems, drivers, firmware tooling, or other open-source systems software.
  • Have delivered or supported system-software changes across large production compute fleets.

Similar open jobs

Software engineer jobs in San Francisco (1709)Jobs in San Francisco (11750)