AI Hardware Architect - Low-Precision & Sparse Compute

Beijing Foreign Enterprise Management Consultants Co.,Ltd.; Singapore; On-site; SGD 180,000 - 260,000; Full time

Job summary

Huawei seeks an Advanced Engineer (High-Efficiency AI Computing) to shape the next generation of AI hardware. You will bridge algorithmic innovation and silicon design, advancing low-precision and sparse computing on in-house AI accelerators and driving energy efficiency and throughput for large-scale AI models.

Responsibilities include leading quantization research, building scalable kernels (GEMM, FlashAttention), and partnering with IC design to deliver technical specs, benchmarks, and patent

Qualifications

• Ph.D. in Computer Science, Electronic Engineering, Automation, or a highly related technical field.
• Deep foundational knowledge in computer architecture, specifically GPU/NPU microarchitecture implementation.
• Proficiency in algorithm research focusing on low-precision quantization and sparsity.
• Strong track record in inference optimization for LLMs or multi-modal models with kernel development.
• Concrete system-level architectural design experience for high-performance AI inference accelerators.
• Robust publication record in tier-1 architecture and AI conferences (ISCA, MICRO, HPCA, ASPLOS, NeurIPS, CVPR).
• Exceptional cross-functional collaboration and communication across algorithm, silicon design, software, testing teams.

Responsibilities

• Spearhead research into optimal quantization methodologies for low-precision data formats.
• Architect and deploy high-performance kernels for low-precision and sparse computing.
• Drive microarchitectural evolution of AI chips to improve energy efficiency and throughput.
• Collaborate with IC design teams to deliver specifications, benchmarks, and patent proposals.

Skills

GPU/NPU microarchitecture

Computer architecture

Low-precision quantization

Sparsity

Inference optimization for LLMs

High-performance kernel development

Cross-functional collaboration

Education

PhD in Computer Science or Electrical/Electronic Engineering

Tools

GEMM kernels

FlashAttention

CUDA

Microarchitecture design tools

Job description

Huawei seeks an Advanced Engineer (High-Efficiency AI Computing) to shape the next generation of AI hardware. You will bridge algorithmic innovation and silicon design, advancing low-precision and sparse computing on in-house AI accelerators and driving energy efficiency and throughput for large-scale AI models.

Responsibilities include leading quantization research, building scalable kernels (GEMM, FlashAttention), and partnering with IC design to deliver technical specs, benchmarks, and patent