I'm an undergraduate at the University of Michigan, Ann Arbor, studying Data Science, after two years in Mechanical Engineering at Shanghai Jiao Tong University. My research spans machine learning and the sciences: building data-preparation and training pipelines for genomic foundation models (e.g. TSS-to-expression prediction), studying how large language models can support incentive design in shared-mobility systems, and developing AI-guided sensing for sparse, dynamic environmental systems such as the Great Lakes.

I like the engineering that makes research reproducible — leakage-safe feature pipelines, HDF5-backed data processing, HPC/Slurm training with bit-equivalent optimizations, and honest statistical validation. Before Michigan I interned at Mars Asia's Global IT Service Center on full-stack WeCom development with JavaScript, Node.js, and Vue.js, building internal tools and automation. I also build interactive web applets for materials-science data analytics.

I'm early in my undergraduate studies and looking for research questions I can go deep on — especially where careful representation and rigorous pipelines turn scientific data into something models can learn from. A fuller list of my work lives on my GitHub.