AI SYSTEM ENGINEER, HIGH-PERFORMANCE COMPUTING (HPC / GPU / NPU) | GLOBALAI INFRA PLATFORM
Our Client is a leading global technology company developingnext-generation AI infrastructure and accelerator platforms. The organisationadvances large-scale AI computing through innovations in reduced-precision and sparsity-driven computing,combined with hardware–software co-optimization,delivering significant gains in performance and energy efficiency.
Reporting to senior technical leadership, this role sits at theintersection of algorithms, architecture, and silicon,driving high-efficiency AI computing. You will play a key role in shaping AI accelerator microarchitecture and optimizingperformance for large-scale AI models (LLMs / multimodal systems).
· Lead research in low-precision quantization and sparse computing algorithms
· Design and optimize high-performance kernels (e.g., GEMM, Attention)
· Drive hardware–softwareco-design to enhance chip performance and efficiency
· Identify system bottlenecks andinfluence next-generation AI chip architecture
· Collaborate with IC teams on specifications, benchmarking, and optimization
Master’s or PhD in Computer Science, Electronic Engineering, or a relatedfield, with a strong foundation in GPU/NPU and computer architecture.Demonstrated expertise in quantization, sparsity, or AI acceleration algorithmsis required, with experience in LLM or multimodal inference optimizationpreferred. Exposure to high-performance kernel optimization or system-level AIaccelerator design is a plus, while publications in top-tier conferences (e.g.,ISCA, MICRO, HPCA, ASPLOS, NeurIPS, CVPR) are advantageous. Strongcross-functional collaboration and communication skills are essential.