Chief xPU Architect for Mobile GPU & AI Convergence

Huawei; Cambridge; On-site; GBP 120,000 - 190,000; Full time

Benefits offered by this job

Annual leave

Pension

Life insurance

Private medical

Medical expense

Employee assistance

Cycle to work

Sports & social

Learning & development

Job summary

Huawei Research and Development UK Limited seeks a seasoned GPU architect to lead the converged xPU architecture for future AI compute. You will shape a unified GPU/NPU design, optimize memory and interconnects, and collaborate across CPU, software and system teams to drive performance, power efficiency and silicon success.

With 15+ years in GPU or heterogeneous compute, you will influence architectural trade-offs, oversee tape-out readiness, and contribute to a next-generation mobile compute

Qualifications

• 15+ years of experience in GPU, AI accelerator, or heterogeneous compute architecture
• Deep understanding of GPU microarchitecture (SIMD/SIMT, scheduling, memory systems)
• Strong knowledge of tensor/matrix computation and AI acceleration techniques
• Proven experience delivering high-volume silicon
• Expertise in performance modelling and power analysis

Responsibilities

• Lead the definition and execution of converged xPU architecture for future AI compute
• Analyse future mobile graphics and AI workloads and redefine a converged GPU/NPU architecture
• Define memory hierarchy and on-chip interconnect strategies for heterogeneous compute
• Collaborate with CPU, AI software, and system teams across SoC development
• Drive performance, power and area efficiency from concept to tape-out

Skills

GPU architecture

System-level architecture

Heterogeneous compute

Graphics & AI workloads

Leadership & collaboration

Job description

Huawei Research and Development UK Limited seeks a seasoned GPU architect to lead the converged xPU architecture for future AI compute. You will shape a unified GPU/NPU design, optimize memory and interconnects, and collaborate across CPU, software and system teams to drive performance, power efficiency and silicon success.

With 15+ years in GPU or heterogeneous compute, you will influence architectural trade-offs, oversee tape-out readiness, and contribute to a next-generation mobile compute