Head of AI Compute Infrastructure

Crux AI; Palo Alto (CA); Hybrid; USD 350,000 - 700,000; Full time

Benefits offered by this job

Health, dental, and vision coverage

401(k) Plan with 4% company match

Hybrid schedule: four days in-office +

Bonus/incentive compensation

Life insurance and disability

Job summary

Crux AI, a U.S.-based integrated AI infrastructure company, is seeking a senior platform leader to own the compute platform strategy and architecture for TPU clusters, fleet provisioning, and a multi-layer engineering organization. You will drive reliability, efficiency, and scale across a cloud-like bare-metal stack.

You will partner with Google on TPU roadmap and lead a team of engineers to deliver customer-ready compute services, with a strong focus on automation, observability, and cost

Qualifications

• 15+ years building and operating large-scale compute or cloud infrastructure platforms.
• Direct ownership of a production compute platform at scale with reliability and customer outcomes.
• Track record of platform rebuilds or transformations while retaining existing customers.

Responsibilities

• Define the compute platform strategy and architecture for dedicated TPU clusters and delivery to customers.
• Architect and deliver the full compute lifecycle from hardware onboarding to customer handoff.
• Design and ship Crux’s container orchestration on bare metal with automated provisioning and upgrades.
• Build and lead a multi-layer engineering organization across provisioning, orchestration, telemetry, and the customer surface.
• Hire leaders and maintain an engineering cadence to ensure quality and on-time delivery.
• Own compute reliability and efficiency end-to-end with defined SLOs and incident handling at fleet scale.
• Collaborate with Google on TPU hardware/software roadmap and translate customer needs into Crux requirements.

Skills

Large-scale compute

Platform leadership

AI-first engineering

Scheduler design

Job description

Crux AI, a U.S.-based integrated AI infrastructure company, is seeking a senior platform leader to own the compute platform strategy and architecture for TPU clusters, fleet provisioning, and a multi-layer engineering organization. You will drive reliability, efficiency, and scale across a cloud-like bare-metal stack.

You will partner with Google on TPU roadmap and lead a team of engineers to deliver customer-ready compute services, with a strong focus on automation, observability, and cost