Hardware System Architect Lead GPU Platform

About Think

Think AI, founded in Riyadh in 2025, focuses on enhancing compute efficiency by integrating hardware and software design. The company aims to redefine AI infrastructure from the silicon up, building efficient compute platforms where hardware and software co-evolve to maximize intelligence. This approach seeks to enable scalable AI operations without compromise or wasted compute.

The Role: Hardware System Architect Lead GPU Platform

Think AI is seeking a Hardware System Architect Lead for GPU Platforms. This full-time role, based in Riyadh, Saudi Arabia, involves owning the architecture of multi-accelerator server platforms and the ILM orchestration platform. The initial focus will be on a high-density, multi-accelerator system designed for mixed-silicon fabrics, supporting NVIDIA, AMD, and Intel accelerators. The role requires taking platforms from block diagram to manufacturing release and setting the direction for future node generations. Over time, this position will build an in-house board design team.

Key Responsibilities

• Own the end-to-end system architecture, including CPU platform selection, PCIe Gen5 and Gen6 topology, switch selection and partitioning, lane budgets, and GPU-to-NIC affinity.
• Design for GPUDirect RDMA and peer-to-peer traffic across heterogeneous accelerators, validating ACS, ATS, IOMMU, and P2P behavior for each silicon vendor.
• Manage the signal integrity strategy across retimers, MCIO, cabled PCIe, risers, backplanes, and channel loss budgets, collaborating with layout and SI engineers.
• Lead power and thermal architecture for dense, high-TDP accelerator configurations, covering CRPS power delivery, airflow, and liquid-cooling readiness.
• Define BIOS/UEFI and BMC requirements, working with the ILM software team on bring-up, telemetry, firmware, and fabric integration.
• Write platform specifications and reference designs for JDM partners, and own design reviews of schematics, layouts, and BOMs.
• Select and manage ODM and JDM partners for motherboards, chassis, and PCIe switches, driving platforms through EVT, DVT, and PVT into production.
• Lead the transition from JDM to in-house board design, taking ownership of schematics, layout, and SI/PI, and define the roadmap for future generations, including Gen6 switching, CXL, and rack-scale integration.
• Own platform validation and quality strategy, including validation plans, qualification matrices, and entry/exit criteria for EVT, DVT, and PVT.
• Drive Design for Test and Manufacturing (DFx) with the manufacturing team to ensure testability at board, server, and rack levels.
• Lead failure analysis and root cause for platform issues from power-on through mass production, developing corrective-action reports and debug guides.
• Build validation and test automation in Python and Bash, integrated with ILM telemetry.
• Build and lead the hardware platform team, including board design, SI/PCIe, mechanical/thermal, platform firmware, and platform validation and test.

Qualifications and Experience

• A minimum of 10 years of experience in hardware system architecture.
• Demonstrated experience in shipping dense multi-GPU platforms.
• Experience with lane budgets, switch partitioning, and bringing systems through EVT, DVT, and PVT into volume production.
• Proficiency in platform quality assurance, including validation at board, server, and rack levels.

Company Values

Think AI operates on four core principles:

• Ownership: Taking responsibility for outcomes, managing resources frugally, and prioritizing the mission.
• Agility: Moving with velocity, shipping quickly, iterating, and committing once decisions are made.
• Impact: Building for the customer, thinking broadly, executing simply, and measuring success by outcomes.
• Mastery: Deeply addressing complex problems, maintaining excellence as a default, and taking pride in customer satisfaction.

Benefits

• Opportunity to work with one of the first deep tech companies in the region, developing foundational technology in-house.
• Meaningful ownership and impact at an early stage.
• Competitive early-stage compensation.
• Close collaboration within a small, senior team.
• Engagement with problems combining hardware, systems, and AI at scale.