Compute Capacity Acquisition

fal•San Francisco, CA

About The Position

fal is a generative media ecosystem providing infrastructure, tools, and model access for AI products. The Compute Capacity Acquisition role ensures fal can obtain the GPUs it needs at the right price and time. This is a technical and commercial role focused on supplier relationships, capacity evaluation, market intelligence, negotiations, sourcing, and forecasting. It is neither a sales role nor a procurement desk role. The role owns fal’s relationships with hyperscalers and neoclouds, including NVIDIA cloud partners, and evaluates capacity offers from GPU specifications through data center infrastructure. It works with fal’s deal team and fleet operations team to translate capacity requirements and demand forecasts into sourcing plans and negotiated agreements. The role also develops a market outlook on supply, GPU platform transitions, and regional or policy changes, and translates that perspective into sourcing decisions.

Requirements

Bachelor’s degree in engineering, computer science, business, economics, or a related field.
At least 5 years of experience in compute or infrastructure capacity, sourcing, partnerships, or strategy at a hyperscaler, neocloud, GPU marketplace, chip vendor, OEM, or AI infrastructure company.
Technical fluency with GPUs, servers, and data center infrastructure sufficient to evaluate supplier offerings.
Commercial judgment, including comfort with unit economics and contract structures and the ability to recognize when a price may indicate a problem.
A track record of building relationships across large, complex organizations.
Willingness to travel to evaluate sites and infrastructure in person.

Nice To Haves

Existing relationships at hyperscalers or major neoclouds, including NVIDIA cloud partners.
Experience managing allocation, forecasting, or partner standing directly with NVIDIA or a tier-1 OEM.

Responsibilities

Build deep, multi-level relationships across hyperscalers and neoclouds, including NVIDIA cloud partners, so fal is a first call when capacity becomes available.
Evaluate capacity offers, including hardware generation, node and rack configuration, interconnect, power density, and cooling, and assess expected performance beyond the supplier’s specifications.
Track prices by GPU generation, region, and contract term; explain price movements and advise when fal should buy, wait, or switch.
Work with the deal team to negotiate capacity commitments, deposits and payment timing, SLAs and remedies, and carry agreements through signature.
Translate GPU requirements from the fleet operations team and demand forecasts into sourcing plans covering suppliers, regions, and terms.
Develop and defend a view on incoming supply, GPU-generation transitions, and regional or policy shifts, and turn it into concrete sourcing decisions.