AI Frameworks Engineer (Model and Kernel Optimization)

Location: On-site in Shanghai, China

Company

AI engineering organization developing software stacks and high-performance deep learning solutions for customer frameworks and Intel accelerators.

What you will do

• Optimize performance for key AI use cases and models.
• Debug and resolve issues involving model accuracy and memory management.
• Design and develop model deployment architectures for accelerated inference.
• Implement new features in vLLM and SGLang, including P/D disaggregation.
• Develop and debug high-performance kernels for Intel accelerators.
• Collaborate with engineers and architects to discuss issues, propose solutions, and gather feedback.

Requirements

• Master's or Ph.D. degree in Computer Science, Artificial Intelligence, Software Engineering, or a related field.
• Strong programming skills in C++ and Python.
• Solid understanding of deep learning fundamentals and hands-on experience.
• Proficient written and spoken English.
• On-site presence in Shanghai is required.
• Strong problem-solving skills and proactive thinking.

Nice to have

• Experience with LLMs or AIGC and understanding of model architectures.
• Experience with PyTorch, vLLM, or SGLang.
• GPU kernel development experience.

Culture & Benefits

• Full-time position for a college graduate.
• Shift 1 schedule in China.
• Collaborative work with teammates, partners, and architects.
• Ethical hiring practices with no recruitment or medical examination fees.