Role Overview

The Frontier Systems team at OpenAI is at the forefront of AI infrastructure innovation. We design, build, launch, and maintain the largest supercomputers on the planet, purpose-built for training OpenAI's frontier AI models. As a Software Engineer on this team, you'll transform data center blueprints into reliable, high-performance systems capable of running uninterrupted training runs for the most advanced AI systems.

Your work directly supports OpenAI's mission to develop safe artificial general intelligence (AGI) that benefits all of humanity. Even a single hardware failure can cost weeks of compute time and millions in resources, so reliability is paramount. Engineers here are trusted with full ownership—from diagnosing complex system issues to deploying automation that scales across thousands of GPUs and nodes.

This isn't just ops work; it's systems engineering at the bleeding edge. You'll dive deep into root causes, build tools that prevent failures proactively, and collaborate with top AI researchers to keep training pipelines humming 24/7. No prior hardware experience required—we'll teach you the low-level details like PCIe protocols, Infiniband networking, and kernel tuning as you go.

Based in San Francisco, this role offers the excitement of working on infrastructure that pushes the boundaries of what's possible in AI compute.

Key Responsibilities

In this high-impact role, you'll take end-to-end ownership of the systems that power frontier model training. Here's what you'll do daily:

• Own system health checks that ensure hyperscale supercomputers remain stable during multi-week training runs.
• Lead investigations into hardware failures, analyzing petabytes of telemetry to pinpoint root causes.
• Build Python-based automation to monitor and remediate issues across 10,000+ machines in real-time.
• Dig into noisy logs and metrics using SQL queries, PromQL, and Pandas for reproducible insights.
• Develop shell scripts and tools for low-level hardware interactions, like power cycling nodes or diagnosing Infiniband link flaps.
• Optimize kernel parameters for peak GPU utilization and minimal latency in distributed training.
• Create dashboards and visualizations for data center-wide health monitoring.
• Collaborate with hardware vendors to resolve systemic issues at exascale.
• Design failover logic to handle node failures without interrupting model training.
• Scale monitoring systems as clusters grow from thousands to tens of thousands of GPUs.
• Document failure modes and build preventive automation to eliminate them.
• Support 24/7 on-call rotations with a focus on minimizing researcher downtime.
• Contribute to open-source tools for large-scale systems management (where possible).

Expect to wear multiple hats: builder, detective, and optimizer—all while shipping code that runs the future of AI.

Qualifications

We're looking for senior engineers who thrive in ambiguity and scale. Required:

• 7+ years of software engineering experience in production environments.
• Expert Python and shell scripting for automation and tooling.
• Proven ability to wrangle noisy data with SQL, PromQL, Pandas, or similar.
• Track record of building reproducible analyses that drive decisions.
• Balanced skills in software development and operational reliability.

Bonus points for:

• Hands-on experience with hardware internals (PCIe, Infiniband, networking).
• Data center visualization and monitoring expertise.
• Network ops, power management, or Linux kernel tuning.

Prior supercomputing or AI infra experience is a plus, but not required. Strong systems thinkers from cloud, HPC, or large-scale web services will excel here.

Salary & Benefits

Competitive total compensation for senior Software Engineers in San Francisco ranges from $250,000 to $450,000 base, plus equity and bonuses. OpenAI offers one of the best packages in tech:

• Top-tier medical, dental, vision coverage.
• 401(k) with 4%+ match.
• Unlimited vacation and flexible hours.
• 16+ weeks parental leave.
• Wellness stipends and gym reimbursements.
• Relocation support for SF move.
• Stock options in a unicorn shaping AGI.

Full details shared during interviews.

Why Join OpenAI?

OpenAI isn't just another tech company—we're building AGI to benefit humanity. Your code will run on supercomputers training models that could solve climate change, cure diseases, and accelerate scientific discovery. Join a mission-driven team of the world's best, with a culture that values impact over bureaucracy.

San Francisco HQ offers vibrant collaboration, stocked kitchens, and events. We're equal opportunity employers committed to diversity.

How to Apply

Submit your resume and a note on why you're excited about Frontier Systems. Interviews include technical deep dives, systems debugging, and team fit. No agencies, please.

Apply now and power the next era of AI!