VP, Compute

Crux AI is a newly formed, U.S.

-based integrated AI infrastructure company created to remove the physical and operational constraints on consequential AI ambitions. Crux brings together power, high-density data centers, TPU silicon, networking, orchestration software, and ongoing operations as one integrated system.

Crux is being capitalized to plan every layer together, develop each one to demanding standards, and operate the whole system with efficiency and reliability. That gives hyperscalers, frontier AI labs, sovereign customers, enterprises, and AI-native companies greater freedom to pursue the AI they are here to create.

You will own compute at Crux: the platform that turns Google TPUs, Crux power and data center capacity, and our network into dedicated, reliable, customer-ready AI compute. Reporting to our CTPO, you own everything between the rack and the customer's first job: the TPU cluster fleet, the software that provisions, monitors and heals it, the container orchestration layer that schedules work onto it, and the compute products our customers consume, from our first phase of capacity through a multi-gigawatt fleet.

This is a platform-builder seat: you will set the architecture of record, hire the leaders, and hold the bar for reliability, efficiency, and speed as Crux scales.

• Define the compute platform strategy and architecture of record: how TPU capacity is carved into dedicated clusters, provisioned, isolated, and delivered to hyperscalers, frontier AI labs, sovereign customers, and enterprises.

• Architect and deliver the full compute lifecycle: hardware onboarding and inventory, OS and firmware provisioning, burn‑in and validation, rack certification, customer handoff, repair and retirement.

• Design and ship Crux's managed container orchestration on bare metal: automated control‑plane provisioning, upgrades, and operators for TPUs.

• Build and lead a multi‑layer engineering organization spanning cluster provisioning and lifecycle, orchestration and scheduling, fleet health and telemetry, and the customer‑facing compute surface.

• Hire people better than you at their craft, then stay close enough to know when something is off before it shows up in a report.

• Own compute reliability and efficiency end to end: SLOs, goodput and utilization, time from rack‑ready to customer‑ready, and failure recovery at fleet scale, achieved through discipline, not heroics.

• Serve as Crux's primary technical counterpart to Google on the TPU hardware, software, and services roadmap, translating customer needs into Google requirements and Google capabilities into Crux product.

• Set the build, adopt, or integrate line for the compute software stack: what Crux builds natively, what it adopts from Google, and what it sources from partners.

• Stand up the security, tenant isolation, and compliance posture that sovereign and regulated customers require for dedicated compute.

• Partner with Data Center Engineering and Capacity Delivery on cluster design, bring‑up, and acceptance; with Network Engineering on fabric and interconnect; with TPM's on what we can commit to and when; and with Finance on the utilization, failure‑rate, and spares inputs to the cost and margin model of record.

• Gained a complete understanding of the TPU platform, the Google partnership model, the capacity plan of record, and our first customer commitments, and published a written compute architecture and org plan.

• Hired or committed your first layer of engineering leaders across the core compute domains.

• Established working cadences with Google counterparts, Data Center Engineering, Network…