Role Overview

The Fleet team at OpenAI manages the computing backbone that drives our cutting-edge AI research and deployment. Spanning massive data centers filled with GPUs, high-speed networking, and custom infrastructure, our systems must deliver flawless performance at unprecedented scale. A single hardware failure can halt multi-billion-parameter model training or disrupt millions of ChatGPT users worldwide.

In this pivotal role, you'll pioneer solutions for state-of-the-art supercomputers that don't exist anywhere else. We're not just operating infrastructure—we're inventing the operational paradigms for AI at exascale. You'll enjoy high autonomy, deep ownership, and the ability to drive meaningful change across OpenAI's compute ecosystem.

Prior hardware expertise isn't required. We value engineers who excel at system-level thinking, automation, and turning chaos into reliable systems. If you thrive on comprehensive investigations, building at scale, and eliminating toil through code, this is your opportunity to shape AI infrastructure's frontier.

Key Responsibilities

Your day-to-day will focus on making OpenAI's compute fleet unbreakable:

• Architect automation systems that provision and manage tens of thousands of GPU servers across global data centers
• Build sophisticated monitoring solutions tracking server health, thermal performance, memory errors, and network fabric integrity
• Lead cross-team collaborations with cluster orchestration, networking, and capacity planning groups
• Work directly with colocation partners to enforce rigorous quality standards and rapid issue resolution
• Profile and eliminate performance bottlenecks across the entire GPU stack—from PCIe lanes to inter-node Infiniband fabrics
• Engineer continuous improvements that slash manual operations by orders of magnitude
• Perform exhaustive root cause analysis on fleet-wide anomalies using every tool in your arsenal
• Develop proactive detection systems that predict failures before they cascade
• Fine-tune Linux kernels and hardware firmware for optimal HPC workloads
• Manage full hardware lifecycles from procurement through decommissioning at supercomputer scale
• Integrate observability across Prometheus, Grafana, custom agents, and ML-powered anomaly detection
• Troubleshoot bleeding-edge systems where you're writing the playbooks as you go

Every decision you make directly impacts OpenAI's ability to push AI boundaries responsibly and reliably.

Qualifications

We're looking for engineers who combine builder and operator DNA:

• Proven experience managing production server fleets at significant scale (10k+ nodes preferred)
• Demonstrated ability to build AND reliably operate mission-critical infrastructure
• Strong proficiency in Python, Go, Rust, or similar for infrastructure automation
• Deep Linux expertise including kernel debugging, networking stack, and system tuning
• Experience wrangling messy telemetry data with SQL, PromQL, Pandas, or equivalent
• Track record of deep-dive investigations that uncover root causes others miss
• Passion for automation that scales linearly while humans scale logarithmically

Bonus points for: Low-level hardware familiarity (PCIe, Infiniband, IPMI/Redfish), HPC experience, distributed systems design, or monitoring tool mastery. But don't hesitate if your background is software-heavy—we teach the hardware side.

Salary & Benefits

Compensation Range: $250,000 - $450,000 USD base salary (San Francisco), plus competitive equity package. Total compensation calibrated to senior infrastructure engineering market rates for top HPC talent.

Comprehensive Benefits Package:

• Market-leading medical, dental, vision coverage
• 401(k) with generous matching
• Unlimited PTO with recharge encouragement
• Generous parental leave (primary + secondary)
• Fertility assistance benefits
• Mental health professional support
• Professional growth stipend
• Relocation support for SF move
• Premium office perks and wellness reimbursement

OpenAI structures compensation to attract and retain exceptional talent building humanity's most important technology.

Why Join OpenAI?

OpenAI isn't just another tech company—we're creating artificial general intelligence to benefit all humanity. Your work directly enables:

• Training next-generation foundation models that redefine what's possible
• Serving hundreds of millions through products like ChatGPT
• Safety-first deployment of increasingly powerful systems
• Research pushing fundamental AI boundaries

Work with brilliant peers who value diverse perspectives. Enjoy high ownership, rapid iteration, and tangible impact. We're building infrastructure that will power AGI, and you can be at the forefront.

OpenAI is an equal opportunity employer committed to diversity across all dimensions.

How to Apply

Ready to safeguard the compute powering AGI? Submit your resume and a brief note about your most impactful infrastructure project. Tell us about a time you automated away months of manual work or solved a fleet-wide mystery. We're excited to hear from builder-operators passionate about reliable systems at scale.

Locations: San Francisco, CA (primary). Exceptional candidates considered for remote within US time zones.

Applications reviewed on rolling basis. OpenAI provides reasonable accommodation during the application process.