Within environmental health, I focus on making association analyses methodologically valid and causally interpretable under the non-ideal data conditions that characterize real-world studies of environmental health effects, including understudied questions in environment, climate, and health. At the core of my work is a commitment to resolving uncertainties across the research pipeline to better inform environmental health decisions, building upon my domain knowledge and real-world experience in environmental health, epidemiology, exposure assessment, and computational algorithms. This work requires the development of integrated frameworks for large-scale data curation/integration, environmental exposure modeling, outcome modeling, bias correction, and the identification of target effect estimands in non-ideal or suboptimal data settings with limited computational infrastructures.

To me, epidemiology, biostatistics, and data science are tools to advance environmental health. My training spans environmental and health sciences, complemented by hands-on experience with in-situ measurements during my undergraduate and postgraduate studies. I rely on core principles in environmental chemistry, physics, and biology, and because many real-world data are inherently imperfect (e.g., measurement error, selection processes, and limited information), I view scientific theory as essential for guiding effective data preprocessing and modeling. I also maintain a personal interest in social science, philosophy, and the use of epidemiology, risk assessment, and environmental health in legal contexts.

I believe that good science requires more than simply applying computational algorithms; it begins with understanding the phenomena we investigate, the data we analyze, the assumptions we make, and the processes that generate them. Much of my work therefore begins before modeling: reading the literature, examining the data, reflecting carefully on study design, and asking whether the mathematical properties of the data and models can support the assumptions needed for causal interpretation. Across different research settings, I have repeatedly seen how seemingly minor or easily overlooked complexities in observational data can meaningfully change scientific findings.

My research spans substantive studies as well as theoretical, methodological, and simulation studies.

I am committed to integrating research, teaching, and mentoring to advance environmental health and train the next generation of scientists.

This commitment is also personal. English is my second language, and I learned to speak and listen only after my mid-20s. As a first-generation college student and the primary breadwinner for my family of three in my early adulthood, I care about capacity building.

My students and trainees develop independent research and participate in a weekly journal club focused on environmental epidemiology, exposure science, prediction, and causal modeling.