Technical Program Manager (TPM), Infrastructure

• Own COGs and R&D attribution programs. Stand up cost attribution across inference, compute, and infrastructure workloads. Partner with Finance and Data to connect spend to products, features, and business metrics. Make cost data actionable for engineering and executive decision-making.
• Drive GPU and infrastructure resource allocation. Build frameworks for capacity planning and allocation across 1P models, 3P inference, and experimentation. Run recurring reviews with ML and Infra leadership. Translate business priorities into concrete allocation plans.
• Manage technical programs with infrastructure teams. Scope and drive programs in cost optimization, capacity planning, reliability, and migration. Own the operating rhythm for each program alongside a designated sponsor: track progress, escalate blockers, ensure outcomes.
• Partner strategically with engineering leaders. Serve as a thought partner on resource tradeoffs, investment prioritization, and operational improvements across ML, Infrastructure, and Finance. Synthesize complex technical and financial information into clear recommendations.
• Roll up your sleeves. Dig into dashboards, data, and systems. Write the doc, build the model, run the analysis, and get stuff done.

Sample projects

• Stand up COGs attribution: mapping inference and infrastructure spend to products and workloads, with dashboards used by engineering and finance.
• Build a GPU allocation review that runs bi-weekly with ML and Infra leadership, producing clear decisions tied to business priorities.
• Drive an R&D spend efficiency program identifying the highest-leverage cost reductions across engineering.
• Scope a capacity forecasting model connecting product growth to infrastructure investment needs for planning.

Who thrives in this role

• Thinks in systems: connects infrastructure decisions to business outcomes.
• Can operate at every level. Equally comfortable in a strategy discussion and a spreadsheet.
• Builds trust with senior technical leaders by being prepared, direct, and reliable.
• Operates with high ownership. Doesn't wait to be told what to do.