Senior Hardware Systems Architect for AI Compute Platforms

Celestica; San Jose (CA); On-site; USD 180,000 - 260,000; Full time

Job summary

Celestica seeks a Principal Hardware Systems Architect for Compute Platforms to lead the architecture and development of AI, accelerated computing, and high-performance data center platforms. You will drive end-to-end compute-system definitions, spanning CPUs, GPUs, memory, storage, PCIe/CXL fabrics, and scale-up interconnects across rack and multi-rack deployments.

You will partner with Network, Power, Thermal, Mechanical, Firmware, and Validation teams to deliver integrated AI systems,

Qualifications

• Bachelor’s or Master’s degree in Electrical Engineering, Computer Engineering, Computer Science, Physics, or related technical field.
• Typically 12+ years of hardware and system-development experience in server/compute/AI/HPC/data-center platforms.
• Demonstrated experience defining or leading system-level architecture for complex compute platforms.
• Strong knowledge of x86 and/or ARM processors, GPUs/accelerators, DDR/HBM, PCIe, CXL, NVMe/storage, high-speed interconnects.
• Understanding relationships among compute, networking, memory, power, and thermal architecture.
• Experience taking complex platforms from architecture through development, validation and production.
• Ability to make and communicate system-level engineering tradeoffs (performance, cost, power, thermal, reliability, schedule).
• Strong technical communication and presentation skills.
• Ability to influence technical decisions across multiple engineering disciplines without direct authority.
• Experience engaging with customers, silicon vendors, and technology partners.

Responsibilities

• Define architecture for next-generation AI, accelerated-compute, server, and HPC platforms.
• Own architectural definition of CPU/GPU/AI accelerator subsystems and interconnects.
• Define memory, storage, PCIe/CXL fabrics, and rack-scale topology.
• Establish compute-system requirements, interfaces, partitioning, and tradeoffs.
• Evaluate alternatives for performance, bandwidth, latency, power, and cost.
• Develop reusable compute-platform architectures and building blocks.

Skills

System-level architecture

CPU architectures

GPU/AI accelerators

Interconnects (PCIe/CXL)

Thermal/power management

Technical leadership

Customer engagement

Engineering communication

Education

Bachelor's or Master’s degree in Electrical Engineering, Computer Engineering, Computer Science, Physics, or related field

Tools

NVLink/NVSwitch technologies

Networking interconnects

Job description

Celestica seeks a Principal Hardware Systems Architect for Compute Platforms to lead the architecture and development of AI, accelerated computing, and high-performance data center platforms. You will drive end-to-end compute-system definitions, spanning CPUs, GPUs, memory, storage, PCIe/CXL fabrics, and scale-up interconnects across rack and multi-rack deployments.

You will partner with Network, Power, Thermal, Mechanical, Firmware, and Validation teams to deliver integrated AI systems,