Hardware System Design Director (AI Accelerator Card Focus)
Founded in 2020, OUR CLIENT is a leading Chinese sparse computing AI chip design company, focusing on the R&D of next-generation AI computing infrastructure for the era of large models. The company pioneered the "Dual Sparsity" technical architecture, significantly improving the inference efficiency and energy efficiency ratio of large models through hardware and software co-optimization, aiming to build a world-leading AI inference chip platform.
As a representative of innovation in China's AI chip industry, OUR CLIENT was selected for EE Times' global "Silicon 100" list in 2024 and won industry honors such as "2024 AIGC Top 100 Innovative Enterprises". The company's products cover cloud inference chips, AI computing cards, and computing cluster solutions, and have been commercialized in scenarios such as cloud computing, large model services, and intelligent customer service.
In terms of financing, the company has completed angel round, Pre-A round, A+ round, and B round financing, with a total financing amount of hundreds of millions of RMB. Investors include renowned institutions such as Triumph Ventures, Jiangmen Ventures, ZhenFund, Jinpu Investment, Huada Songhe, and Ant Group.
Currently, OUR CLIENT is accelerating the mass production and commercialization of its large- model inference chips, building a complete ecosystem covering "algorithm—chip— computing power platform—industry applications," and has become one of the most promising innovative companies in China's AI computing power infrastructure field.
Hardware System Design Director (AI Accelerator Card Focus)
Work Location: (To be determined)
Reporting to: CTO / Hardware R&D Manager
Job Responsibilities:
1. Responsible for defining AI accelerator card product specifications and overall hardware architecture design; leading accelerator card PCB development; coordinating structural design, heat dissipation planning, and product implementation.
2. Responsible for the hardware architecture design and R&D of AI accelerator cards and high-speed interconnect products; proficient in high-speed interface design such as PCIe,NVLink, CXL, and Ethernet; solving key technical issues such as signal integrity (SI) and power integrity (PI).
3. Responsible for hardware adaptation, compatibility debugging, and system integration testing of AI server and Super Node platforms, driving product verification, certification, and mass production.
4. Participate in server hardware platform selection, technology roadmap planning, and product iteration; formulate hardware design specifications and development processes.
5. Coordinate teams involved in hardware, architecture, thermal management, testing, and supply chain to ensure projects are delivered on schedule.
6. Tracking the development trends of GPU, AI computing, high-speed interconnect, and data center hardware technologies, and promoting the application of cutting-edge technologies in products.
Job Requirements:
1. Bachelor's degree or above in Electrical Engineering, Microelectronics, Communication Engineering, Computer Hardware, or related fields.
2. 8+ years of hardware R&D experience, with experience in developing AI accelerator cards, GPU servers, high-performance computing (HPC), or high-speed interconnect products.
3. Familiar with high-speed PCB design, high-speed signal link design and debugging, and possess SI/PI analysis and optimization capabilities.
4. Familiar with AI server hardware architecture and understand GPU, CPU, switching chip, storage, and network system design.
5. Experience in AI accelerator cards, server platform adaptation, and system-level verification, with experience in mass production project implementation is preferred.
6. Familiar with structural design, thermal design, and reliability design processes, and able to drive cross-domain collaborative development.
7. Excellent problem-solving, project management, and cross-departmental communication and coordination skills are required.
Preferred candidates will have:
Experience in developing NVIDIA GPU servers, AI training clusters, and high-speed switching network products;
Experience in PCIe Gen5/Gen6, CXL, InfiniBand, and 800G networks;
Have experience as a hardware lead or technical leader with team management skills;
Background in AI chip, server, data center, cloud computing, or high-speed interconnect related companies.
Compensation and Benefits
Competitive salary package + Year-end bonus + Equity/option incentives + Project performance bonus
Five social insurances and one housing fund
Paid annual leave
Meal allowance and employee benefits
Broad career development opportunities
