Global TPU Capacity Architect: Scale & Orchestrate

Baseten; United States; Remote; USD 180,000 - 240,000; Full time

Job summary

Baseten seeks a Global Capacity Manager focused on TPUs to lead the TPU pod fleet lifecycle, secure large‑scale allocations, and build automation for uptime across multi‑cloud environments.

You will work at Baseten to optimize capacity planning, orchestrate workloads, and partner with cloud teams to ensure cost‑effective, reliable TPU infrastructure for customers and internal workloads.

Qualifications

• Experience leading specialized TPU pod fleets and managing lifecycle from acquisition to maintenance.
• Strong skills in advanced orchestration and workload migrations with regional/compliance requirements.
• Ability to design scalable capacity management systems to grow TPU/compute capacity safely.
• Experience with financial modeling related to asset utilization and cost control.

Responsibilities

• Lead Specialized Pods: act as the lead for TPU pod fleets and manage acquisition, allocation, and maintenance.
• Advanced Orchestration: migrate workloads and drain deployments across TPU topologies while meeting regulatory requirements.
• Build for Scalability: design Baseten's next capacity management system to handle TPU growth with existing GPUs.
• Financial Modeling: model capacity and economics to optimize spend and reliability.

Skills

TPU capacity management

Kubernetes orchestration

Cloud infrastructure

Cost optimization

Tools

Google Cloud

Kubernetes

Job description

Baseten seeks a Global Capacity Manager focused on TPUs to lead the TPU pod fleet lifecycle, secure large‑scale allocations, and build automation for uptime across multi‑cloud environments.

You will work at Baseten to optimize capacity planning, orchestrate workloads, and partner with cloud teams to ensure cost‑effective, reliable TPU infrastructure for customers and internal workloads.