Compute Infrastructure Strategy Lead

OpenAI • San Francisco, CA

92d

About The Position

Your charter: deliver step-function improvements in cost, capability, capacity, reliability, and time-to-ready across OpenAI’s infrastructure and compute platform. You’ll take on high-leverage initiatives spanning compute, storage, networking, and systems—cut through complexity quickly, pinpoint leverage points, make clear calls, and drive the changes needed to unlock scale. You’ll work fluidly across teams—engineering, capacity planning, research, infra, product, finance, and business development—to align the technical, operational, and commercial pieces needed to make these bets real. Some workstreams will be strategic and industry-shifting; others will be tactical so we can ship systems sooner than otherwise possible. Near-term focus may include optimizing storage direction and rollout, improving accelerator utilization, advancing interconnect approaches to lower latency and cost, and accelerating site readiness.

Requirements

• Have deep technical experience (infrastructure / systems engineering, TPM, or product) and move comfortably between system detail and program execution.
• Have led ambiguous, cross-functional infrastructure work to tangible outcomes without waiting for perfect information.
• Communicate clearly, align partners, and make sound calls on cost, performance, and operability.
• Prefer measured proofs and execution over slideware and enjoy switching between strategy and hands-on work to fully own projects end-end.

Responsibilities

• Own cross-stack programs end-to-end: define the problem, design the approach, prove it quickly, and carry the result into production.
• Drive rapid clarity on goals, criteria, risks, metrics for complex technical problems without adding heavy process.
• Work closely with engineering, operations, and (when needed) finance, research, product, and external vendors to align decisions and deliver useful infrastructure at massive scale.
• Own outcomes end-to-end, and move fast to validate approaches—using bounded pilots when useful—while balancing total cost, performance, and operability.
• Collaborate with engineering to incubate and execute on the technical decisions that unlock scale.

Benefits

• Relocation assistance offered to new employees.
• Hybrid work model of three days in the office per week.

What This Job Offers

• Number of Employees: 1,001-5,000 employees