Zining He

Hi! I’m Zining He (何子宁), a Computer Science undergraduate at Shanghai Jiao Tong University.

My research interests lie at the intersection of AI systems, formal verification, and large language models. I’m passionate about building reliable and efficient AI systems through rigorous testing, verification, and optimization techniques.

Research Experience

I’m currently working on several exciting projects:

• SpecBridge: Developing a reconstruction-guided pipeline that translates natural-language requirements into Lean formal specifications (NeurIPS 2026 submission, third author)
• Cross-Architecture Operator Testing: Building differential testing frameworks for GPU/NPU neural network operators
• MoE Acceleration: Optimizing mixture-of-experts model inference through adaptive expert allocation

Technical Skills

I work with C++, Rust, Python, and Java, with a focus on systems-level programming and AI infrastructure. My experience spans PyTorch, CUDA, formal verification tools like Lean, and various aspects of computer architecture and operating systems.

Feel free to check out my CV or reach out via email if you’d like to discuss research collaborations or interesting problems in AI systems!