The Role:

We are an AI company building reliable, high-performance model serving infrastructure. Our customers run mission-critical workloads and expect consistency, cost transparency, and predictable scaling. Capacity — GPU, network, and cloud economics — is existential for us. You will be one of the earliest hires focused on making sure we always have the right compute, at the right price, ahead of demand.

Key Responsibilities:

• Architect the Multi-Year Capacity Strategy: Develop and own the strategic, multi-year capacity plan by synthesizing inputs from the company strategy, business forecast, Industry trend and Engineering’s product roadmap.
• Infrastructure Cost Management: Directly manage the biggest cost on our P&L. Translate market insights, technology advancement, and forward-looking plans into rigorous financial models to enable fast business growth while minimizing Total Cost of Ownership (TCO) per unit of performance (e.g., TFLOP). Partner with infra, finance, and GTM to codify capacity strategy into budgets and KPIs
• Build market intelligence with advanced compute technology roadmap, lead time, and pricing for relevant SKUs (H200/B200/B300/GB200/GB300/MI355, networking, storage)
• Own end-to-end procurement of cloud capacity (GPUs, storage, networking, etc) across multiple vendors (cloud, bare-metal, colocation, integrators, brokers) including RFQs/RFPs, commercial Negotiation. 
• Define Capacity Management discipline with processes and tools: inventory, allocation, unit economics, cost attribution and optimization
• Stand up processes for disciplined renewals, rev-share allocations, and hedging against supply shocks

Minimum Qualifications:

• 5–10+ years of experience in capacity management, sourcing, data center supply, cloud procurement, infrastructure operations, or a related field (e.g., hyperscaler, colocation provider, OEM, or LLM infrastructure startup)
• Demonstrated ability to develop and communicate multi-year capacity plans that align business, product, and financial objectives
• Strong knowledge of GPU and server SKUs, networking topologies, power and space constraints, and a high-level understanding of the global AI compute supply and demand landscape
• Proven experience negotiating contracts valued at seven figures or higher, with a solid grasp of key terms such as commit profiles, flexibility clauses, drawdowns, credits, and SLA remedies
• Proficiency in unit economics and scenario modeling, including total cost of ownership (TCO) and TFLOP-month analysis
• Ability to operate effectively in a fast-paced, low-process environment and contribute to company-defining initiatives

Preferred Qualifications:

• Established network across major capacity suppliers, including NVIDIA partners, cloud providers, brokers, integrators, and colocation vendors
• Experience connecting capacity planning with real-world model serving workload patterns and performance requirements
• Background in AI infrastructure environments, such as hyperscaler clouds or AI infrastructure startups