← Back to jobs

Remote job

Chief Architect, GPU Infrastructure & Rack-Scale Systems - Israel

Other Full-time Permanent Israel

Job details

Not specified Salary
Israel Eligibility
Executive Experience
Full-time Employment

About this role

Role overview

Define the architecture for GPU infrastructure spanning data centers, racks, servers, boards, chips, and dies. The role connects workload needs to detailed hardware requirements and aligns engineering disciplines around systems designed for performance, efficiency, reliability, and deployability.

Responsibilities

- Translate product and workload needs into GPU infrastructure architectures and hardware specifications. - Set requirements and interfaces across racks, servers, boards, chips, and dies with domain specialists. - Architect accelerator and host connectivity, including scale-up links and data center networking. - Define compute, memory, and storage hierarchies based on workload bandwidth, latency, capacity, and data movement. - Establish power, cooling, physical integration, provisioning, orchestration, telemetry, fault management, and serviceability requirements. - Lead tradeoff reviews, align silicon, hardware, networking, firmware, software, and infrastructure teams, and validate assumptions through models, prototypes, and measurements.

Requirements

- Extensive data center infrastructure architecture experience, including deep expertise in GPU or accelerator systems and rack-scale design. - Record of leading complex hardware platforms from architecture through deployment. - Ability to reason from workload and fleet needs down to chip and die constraints. - Strong knowledge of accelerator interconnects, network topologies, memory, and storage systems and their effect on application performance. - Experience with dense power and cooling, system management, reliability, and operational needs. - Technical leadership across organizational boundaries, sound tradeoff judgment, and the ability to turn goals into actionable specifications.

Nice to have

- Experience with large-scale AI training or inference infrastructure. - Familiarity with HBM, DDR, PCIe/CXL, high-speed Ethernet, InfiniBand, accelerator fabrics, chiplets, die-to-die links, or advanced packaging. - Understanding of infrastructure utilization, economics, and total cost of ownership.

Benefits and work setup

- The role offers substantial ownership and the chance to shape foundational architecture. Listed benefits include health coverage, retirement-plan matching, unlimited paid time off, and meals at the Mountain View office.

Detected Nov 17, 2026
Last verified Nov 17, 2026

Hidden Jobs Access

Unlock application links

Read the full job details for free. An active Hidden Jobs Access subscription is required to open the original application link.

Weekly

$6.99 /week
  • Original application links
  • Full access to all MCP features
  • Cancel anytime

Monthly

$35.99 $17.99 /month
  • Original application links
  • Full access to all MCP features
  • 35% cheaper than weekly

Lifetime

$99.99 $49.99 /forever
  • Original application links
  • Full access to all MCP features
  • One-time payment
Hidden Jobs gives subscribers direct access to original application links