Postdoctoral Fellow (AI Infrastructure and Systems for Next-generation AI Chips)

The Department of Information Engineering is seeking Postdoctoral Fellow(s) in the MarsLab-CUHK AI Infra Joint Research Project. The appointee(s) will conduct research on AI systems and software optimization for next-generation AI chip architectures. The project is built upon a mature cloud AI chip platform with completed fabrication, providing a real-world hardware and software research environment for advancing AI infrastructure and system innovation. The research areas include but are not limited to: (1) Large-scale AI model training and inference systems; (2) AI compiler and automatic optimization techniques; (3) High-performance operators, kernels, and runtime systems; and (4) Hardware-software co-design and optimization for emerging AI accelerator architectures.

Applicants should have (i) a Ph.D. degree in Computer Science, Electronic Engineering, Computer Engineering, or related disciplines; (ii) research experience in one or more of the following areas: AI systems, large language model training/inference, AI compiler, high-performance computing, GPU/AI accelerator architecture, or system optimization; and (iii) the ability to conduct independent research and publish high-quality academic papers. Experience with large-scale model systems, distributed training/inference frameworks, compiler optimization, kernel development, or hardware-software co-design will be highly desirable.

The appointee(s) will be responsible for (a) conducting research on AI infrastructure and system optimization; (b) designing and developing efficient algorithms, software frameworks, and system components for AI workloads; (c) publishing research results in leading conferences and journals; (d) collaborating with researchers from CUHK and external project partners; and (e) performing other research-related duties as assigned.

Appointment will initially be made on contract basis for two years, renewable subject to mutual agreement.

The University only accepts and considers applications submitted online for the post above via the CUHK career site. For more information and to apply online, please visit http://career.cuhk.edu.hk.