Technical Program Manager (GPU Compute)

Quick Facts

• Focus: GPU compute planning, capacity forecasting, provider onboarding, and fleet operations
• Partners: Finance, engineering, research, and cloud/neocloud providers

Description

Own how Perplexity plans, brings online, and runs the GPU compute behind products. You’ll forecast multi-month GPU demand, turn it into commitments and budgets, onboard new capacity across providers, and drive improvements in fleet uptime, utilization, and cost. You’ll also build a self-serve operating path for inference and compute requests end to end.

Responsibilities

• Partner with Finance on 3–12 month compute demand forecasting driven by customer growth, model releases, and evaluation plans
• Convert forecasts into decisions with Finance and engineering leadership (commitments and budget) before committing capacity
• Lead transitions for new capacity: provider onboarding, acceptance criteria, migration, and decommissioning
• Build and maintain a single trusted view of the fleet (capacity, cost, allocation, uptime, reliability, utilization by provider and workload)
• Drive daily utilization, reliability, and cost-efficiency; reclaim idle capacity and surface trade-offs
• Own inference and compute requests end to end: triage/resolve GPU, deployment, and quota requests; build self-serve request paths
• Manage compute provider relationships using scorecards and regular reliability/support/delivery reviews
• Partner with engineering and research on capacity decisions for training, batch inference, and real-time inference
• Build the “operating system” for compute: processes and tooling to request, plan, allocate, and track compute with clear owners

Requirements

• 6+ years of technical program management, capacity planning, or infrastructure operations in cloud environments
• Experience leading programs across teams and vendors
• Direct experience with GPU/accelerator fleets across training and inference workloads
• Experience working with cloud or neocloud providers on reserved capacity, commitments, onboarding, or vendor performance management
• Experience with infrastructure cost modeling, unit economics, or cost attribution with Finance
• Technical depth in cluster management, scheduling, and orchestration, plus data skills to build forecasts and dashboards
• Judgment to surface hard trade-offs and communicate them to executives
• Thrives in a small, agile team; initiative and ownership with limited precedent

Benefits

Not specified in the provided posting.