Summary
Yuhao Wang is an applied scientist with eight years of research and industry experience focused on reliable machine learning, statistical inference, and decision-making from biased or partially observed data. Currently at Amazon, he develops causality and outlier-calibration methods that translate theoretical frameworks into product-relevant insights for A/B testing and customer behavior analysis. His background spans academic visits and research roles at MBZUAI, UC Berkeley’s Simons Institute, NUS, and Eindhoven, where he worked on probabilistic graphical models, causality, and edge-aware deep learning systems. Yuhao combines strong statistical rigor with practical skills in SQL and experimental design, enabling robust offline evaluation for real-world systems. He often tackles problems like truncation, censoring, and unmeasured confounding that are overlooked in typical ML pipelines, aiming to make data-driven decisions defensible under imperfect observation.
8 years of coding experience
5 years of employment as a software developer