Head of GPU Capacity Planning

Our mission is simple: deliver seamless, secure, reliable, and resilient AI compute at scale. We've built a versatile cloud platform that eliminates infrastructure barriers, empowering builders to focus on innovation instead of fighting their stack. Because breakthrough AI should move at the speed of ideas, not infrastructure.

TensorWave operates AMD Instinct GPU clusters across multiple U.S. data center sites. As we scale, the ability to see, forecast, and commit capacity with confidence is a core competitive advantage. The Head of GPU Capacity Planning owns that picture.

This person elevates our existing capacity planning processes, drives capacity planning end to end, and maintains a single, authoritative view of the fleet: what we have, what is committed, what is held in reserve, and what is available to sell. They partner closely with Sales to guide sellable capacity, work across active deals to keep commitments feasible, and establish a weekly cadence so capacity is tracked and defended every week rather than reconstructed after the fact. They also translate operating pain into clear tooling requirements that improve how the whole company plans capacity.

If remote, candidate would need to travel to Vegas on a monthly basis.

• →

  Maintain a single source of truth for total, committed, reserved, and available-to-sell capacity across every site and GPU generation, reconciling physical fleet inventory (installed, in burn-in, held as spares, in RMA) against logical and contracted allocations.

• →

  Standardize how capacity is defined, measured, and reported so definitions are consistent and every stakeholder is reading from the same numbers.

• →

  Own demand pipeline and track contracted ramp on curves, steady state consumption, expiries, renewal probability, and take or pay floors, with weighted pipeline maintained separately and never blended into the committed view.

• →

  Define and publish sellable capacity by site, GPU type, and timeframe, and give Sales a clear, current view of what can be committed and when.

• →

  Partner with Sales and Deal Desk across active deals to validate feasibility before commitments are made, flagging oversubscription risk and allocation conflicts early.

• →

  Own the multi-year capacity strategy and act as the connective tissue between Infrastructure Operations, Sales, Finance, and leadership, ensuring the long-range plan reflects real fleet capability.

• →

  Anticipate where capacity constraints will emerge quarters ahead, and drive the cross-functional decisions (deployments, reservations, site expansion) needed to stay ahead of demand.

• →

  Own the buffer and oversubscription policy in partnership with Operations and Finance, and keep those rules visible and enforced.

• →

  Translate the sales pipeline and contracted growth into a rolling capacity forecast, identifying shortfalls and the lead time needed to close them.

• →

  Coordinate with Infrastructure Operations, Global Operations, and supply chain on incoming capacity (racks, nodes, power, cooling, network) so delivery timelines align with committed and forecast demand.

• →

  Act as the connective tissue for capacity decisions, connecting Ops, Sales, Finance, PMO, and the data center teams around one plan.

• →

  Run the weekly capacity review and produce the weekly capacity report and dashboard for leadership, Sales, and Finance.

• →

  Track committed versus available capacity every week with an auditable record of allocations, changes, and the decisions behind them.

• →

  Assess and elevate current capacity planning processes, document the standards and workflows, and raise the maturity of how we plan.

• →

  Define and prioritize tooling requirements for capacity, inventory, and allocation systems (DCIM, dashboards, tracking), and partner with IT, Engineering, and vendors on a build-versus-buy path.

• 5+ years in capacity planning, supply and demand planning, S&OP, technical program management, or infrastructure operations, ideally in cloud, data center, or hardware-intensive environments.

• Working knowledge of data center and compute infrastructure fundamentals: racks, power and cooling constraints, servers and GPUs, and networking, with the ability to reason about both physical and logical capacity.

• Advanced modeling skills and comfort building forecasting and allocation models; strong data fluency (SQL and BI tools a plus).

• A track record of working across Sales, Finance, and Operations, translating technical constraints into commercial guidance that others can act on.

• Excellent written communication and the discipline to run a recurring operating cadence and produce clear, trusted reporting.

• Experience in GPU cloud, HPC, hyperscale, colocation, or semiconductor capacity environments.

• Familiarity with Slurm and Kubernetes cluster capacity, GPU fleet management, and utilization metrics.

• Experience writing requirements for and standing up DCIM, capacity, or inventory tooling.

• Program management certification (PMP or equivalent).

TensorWave is an Equal Opportunity Employer. We celebrate diversity and are committed to creating an inclusive environment for all employees. We do not discriminate on the basis of any protected status under applicable law.

TensorWave provides reasonable accommodations in accordance with applicable laws. If you require accommodation during the hiring process, please contact accomodations@tensorwave.com.

All offers of employment are contingent upon verification of identity and authorization to work in the United States, as required by law.

Where permitted by law, employment may be contingent upon the successful completion of a job-related background check.

By submitting an application, you acknowledge that TensorWave may collect, use, and retain your personal information for recruiting and employment-related purposes in accordance with applicable data privacy laws.