Technical Program Manager, Core Network & WAN Infrastructure

About the role

OpenAI's ability to train frontier models depends on enormous compute clusters, and those clusters depend on a wide area network that can move data at extreme scale. This role sits in the Technical Program Management organization and is responsible for overseeing the complete delivery of OpenAI's WAN infrastructure. You will work hand in hand with engineers to bring new compute clusters online through external partners, and the position is explicitly hands-on and focused on infrastructure execution rather than administrative program management.

a role for someone who understands the physics and logistics of large-scale networking as much as they understand project management. You will be managing the bring-up of large GPU supercomputing clusters across a network of external providers, coordinating everything from fiber and optics to cabling, routing, power, and operational readiness. If a new cluster must go live in months rather than years, you are the person who makes that happen, and who builds the systems that make future bring-ups faster and more predictable.

Key facts

What you'll do

You will lead the full lifecycle delivery of OpenAI's WAN expansion and large-scale GPU clusters across a network of external providers, owning plans, dependencies, and critical paths from initiation through production readiness. You will manage multifaceted network infrastructure bring-up programs that cover both physical and logical readiness, and you will work with engineering teams to activate significant network capacity, aiming to reduce the time it takes to stand up large GPU supercomputers by coordinating across optics and circuits.

A key part of the job is identifying recurring bottlenecks in WAN and network infrastructure deployment and implementing solutions that make future builds faster, more predictable, and less dependent on informal knowledge held by specific individuals. You will coordinate readiness across internal functions including security, finance, operations, and product and research teams, making sure compute capacity is genuinely production-ready when it goes live. You will oversee integrations and handoffs between teams and partners, ensuring consistent execution, clear communication, and quick issue resolution. Where systemic gaps exist, you will drive lasting improvements in tooling, processes, and partner interfaces. And you will keep executives informed with clear updates on progress, trade-offs, and risk across a portfolio of ongoing programs.

Requirements

A degree in a hard science or a proven background in engineering is expected, combined with at least five years of experience in program management for significant projects, including capital projects or hyperscale infrastructure deployment. You need to be resourceful and effective in ambiguous, fast-paced environments, and able to independently lead and deliver complex projects without constant supervision. The role requires a strong technical understanding of physical networking, WAN and backbone infrastructure, colocation environments, cloud interconnects, cross-connects, optics, cabling, routing, and operational readiness. You should also have experience interacting with and managing external vendors such as engineering firms, equipment suppliers, and construction companies, plus expertise in designing straightforward, scalable processes to resolve complex issues and experience managing intricate dependencies, including logistics and supply chains.

Nice to have

• Prior experience at a hyperscaler or cloud provider delivering network capacity.
• Familiarity with GPU cluster architectures and their power and cooling requirements.
• Experience with subsea or metro fiber network buildouts.
• Background in capital project delivery with construction and civil works.

Skills & tools

• WAN and backbone infrastructure
• Colocation and cloud interconnects
• Optics, cabling, and cross-connects
• Routing and network operations
• Program and dependency management
• Vendor and partner management
• Process design and tooling improvement

Practical notes

This role is based in San Francisco, CA, with a hybrid work model requiring 3 days in the office per week. Relocation assistance is available. OpenAI is an equal opportunity employer and provides reasonable accommodations to applicants with disabilities. Background checks are administered in accordance with applicable laws, including the San Francisco Fair Chance Ordinance.

If you are the kind of program leader who cares about the difference between a fiber run and a router config, and who measures success in months saved on bring-up schedules, this is a uniquely impactful role. You will work on the literal backbone of frontier AI infrastructure, and the systems you build will be reused for years as OpenAI continues to scale.