I use human genetics as nature's randomized trial. My work combines Mendelian randomization, evidence synthesis and machine learning to find out which risk factors truly cause disease, and I build software that makes these methods rigorous and reproducible.
I trained first as a civil engineer, optimising earthquake-resistant structures with soft computing, and then earned an M.Sc. in Artificial Intelligence in Taiwan with a perfect GPA and the top rank in my cohort. Prediction alone left me with a question it could not answer: what would happen if we intervened? Since 2023 I have answered that question with genetic data, through Mendelian randomization studies of nutritional, metabolic and reproductive health. In that work I combine statistical genetics with software engineering, and I now want to bring both to East Asian biobank data.
Two-sample, multivariable and drug-target MR; robust estimators for pleiotropy; colocalisation; genetic correlation.
Graph neural networks, attention and reinforcement learning (M.Sc. thesis); interpretable ML and double ML for health data.
Validated, tested tools such as MR Studio and BioCalX that put rigorous methods in clinicians' hands.
Results of unpublished studies are withheld until peer-reviewed publication. Click any card for design and methods.
Indexed in PubMed. My name is shown in bold.
Data Science & AI Applications. Admitted on faculty recommendation as a dual-degree student.
GPA 4.00/4.00 · average 96.46 · ranked 1st of 15. Thesis: Improving autonomous-vehicle trajectory prediction with an enhanced DenseTNT (advisor: Assoc. Prof. Kuei-Chung Chang).
Thesis: optimal seismic design of steel structures with soil–structure interaction using soft computing.
Lead analyst and developer on MR pipelines; Persian translation of Burgess & Thompson, Mendelian Randomization (2nd ed.), with the author's permission.
Trajectory prediction, knowledge distillation, multi-modal gesture recognition.
The only international student inducted at YunTech that year.
Ranked 1st of 15 with a 4.00/4.00 GPA, YunTech.
Selected first-rank for a summer research-lab program.
Selected first-rank for a summer research-lab program.
I welcome conversations about Mendelian randomization, statistical genetics and causal machine learning, especially with groups working on East Asian biobank data.