Principal Product Manager, Compute Platform - Cloud Infrastructure

Overview

Lead product strategy and execution for Netflix's Compute Platform as it evolves into a common, secure, and efficient platform for services, batch jobs, training, inference, and agents at Netflix scale.

What you'll do

• Set the strategic direction for Netflix's Compute platforms in an agent-first world, including guiding the move from Titus to a modern Kubernetes stack and weighing migration costs against the value of a common platform.
• Own the roadmap for GPU capacity provisioning, scheduling across regions, and fleet utilization so training and inference workloads get capacity within seconds while using the fleet efficiently.
• Lead development of a secure, managed environment that supports millions of agents, with built-in identity, isolation, governance, observability, and cost tracking.
• Make compute easier to use by creating a seamless experience across the full compute spectrum, from VMs to functions, guided by regular engagement with service owners, ML engineers, and teams building agents.
• Improve efficiency across the fleet by defining systems that manage compute supply and demand so compute costs grow more slowly than demand, and establish measures for adoption, utilization, cost per workload, and reliability.
• Partner with teams across Ads, Live, Games, and Studio to ensure the compute platform supports their most demanding workloads, including traffic spikes during ad breaks and live events and latency-sensitive processing in regions with limited local infrastructure.

What you'll need

• More than 10 years of experience in technical product management or a similar technical role in compute infrastructure.
• Comfort debating architecture, trade-offs, and system quality with senior engineers.
• Deep knowledge of compute infrastructure, including containers, Kubernetes, scheduling and orchestration, autoscaling, capacity management, multi-tenancy, and cloud economics.
• A strong understanding of AI infrastructure, including GPU provisioning, scheduling, and utilization for distributed training and inference.
• Understanding of agentic workload needs such as secure execution, workload identity, isolation, and cost attribution.
• Experience leading a large replatform or migration from start to finish, including sequencing, adoption, incentives, and the long tail of migration.
• Ability to earn the trust of experienced engineers, challenge their thinking, make clear decisions with incomplete information, and change course when evidence calls for it.
• Strong leadership and communication skills, able to explain complex decisions to technical and executive audiences, build alignment across teams, and influence without relying on authority.

Details

• Location: USA, remote.
• Compensation range: $660,000.00 - $1,000,000.00, varying by location.