- Know GPU infrastructure at scale and can talk about specific systems by generation: A100, H100, H200, B200, B300, and what is different about each in deployment, performance, power, and useful life.
- Built or run large GPU fleets, or worked closely enough with them to know where the cost, the risk, and the failure modes live: networking, scheduling, utilization, and hardware failure.
- Worked on the commercial or financial side of compute: capacity contracts, cost and depreciation models, equipment financing or leasing, or investment diligence on AI infrastructure.
- Explained technical detail in financial terms to people who make decisions on it: investors, lenders, underwriters, or executives.
- Led teams and cross-functional programs. Twelve or more years of experience, with at least five leading teams or leading leaders.
What makes you unusual
Most people in this market are one of two things: an engineer who has never had to justify a dollar, or a financier who has never run a node. We want the person who is credible in both rooms. If you have ever looked at a deal and known within a minute whether the compute behind it was sound, and could then explain why to a room of investors, you are the person we are describing.
Useful but not required: time at a hyperscaler, GPU cloud, or datacenter operator; experience in infrastructure investing, equipment finance, project finance, or insurance; time at a national lab or HPC center.
How we will evaluate you
We will give you a real compute opportunity to take apart: the hardware, the operator, and the contract. We will ask you what it is worth, what could go wrong, and how you would build a solution around it. We care far more about the quality of those answers than about pedigree.
Pay: $175,000 - 275,000
Equity: $100,000 - 200,000 at our latest valuation upon start date