Hello, I'm Zhenhuan Wei.

My current research focuses on AI Infra, training and inference optimization for large language models, with a particular emphasis on distributed systems.

I will soon join Alibaba T-Head, where I will work on framework adaptation and performance optimization for large language models on domestic PPU chips.

Previously, I interned at ByteDance in the Data-AML team, where I contributed to the research and development of a large model framework serving ByteDance's internal search and advertising businesses. Before that, I interned at Tencent IEG, focusing on inference optimization for a TTS large model.

Feel free to reach out: zhenhuan2002@163.com