Telling Stories with Data: A Social Scientist by Day, a Data Scientist Always
I re-search humans, politically, biomedically, and environmentally, using cutting edge computer science tools!

Deep Learning and Gen-AI for Political Science

While large language models excel at text-based applications, they still have a long way to go with images, audio, and structured data such as surveys. I develop deep learning algorithms and adapt computer science paradigms for survey research, directly applying them to my own research in political science. In my dissertation, Three Essays on AI and Political Behavior developed Survey Transfer Learning, which applies the transfer learning paradigm to survey data to generate synthetic samples that are cheaper, more accurate, and more environmentally friendly than current cutting-edge LLM-based silicon samples. This paper, currently under review at Political Analysis, was honored with the national Bud Roper Fellow Award by the American Association for Public Opinion Research (AAPOR). This work, which is the first to apply transfer learning to survey data by leveraging rich available American public opinion data, was selected for presentation at the DC AAPOR and Washington Statistical Society's joint preview conference at Gallup

I first-authored a chapter in Digital Transformation: Accelerating Organizational Intelligence, which examines how digital transformation is reshaping government and public administration. The chapter analyzes how governments leverage emerging technologies, artificial intelligence, knowledge management practices, and organizational restructuring to enhance efficiency, improve decision-making, and strengthen public trust.

Environmental Policy and Politics

My environmental policy research focuses on the IPCC’s call to measure national variation in adaptation, which is a critical part of addressing global warming. Because climate change is a tragedy of the commons, governments under the UN framework are central actors, and I have worked on a grant that codes Nationally Determined Contributions after the Paris Agreement. I presented this paper at the Southern Political Science Association annual meeting, where I also chaired a panel on environmental policy, and it was later presented at APSA in Vancouver. The study demonstrates how reactive and proactive adaptation strategies have evolved and highlights important differences among low-income states, with implications for international adaptation policy. In a related project, I developed a Neoliberal Policy Index using World Bank WDI data, IMF conditionality measures, the KOF globalization index, and labor law indicators to capture privatization, financial openness, and labor deregulation. I found that from 1989 to 2014, countries, especially in Africa, most exposed to neoliberal reforms adopted a paradoxical adaptation strategy: prioritizing donor-aligned adaptations such as refugee shelters over structural ones like agricultural plans.

Health, AI, Evolutionary Biology, and other stories

In addition to my contributions to social science, I am advancing the application of AI in biomedical and evolutionary research. My work on an NIH-funded project, published in the Nature group, explores the function of a biological thermocouple in the devil worm mitochondria under stress conditions, which mirrors cancer cells. I have also contributed to computer vision projects analyzing first of kind long video of organoid formation —an emerging frontier in medical and stem cell research. This research underscores the transformative potential of AI in understanding cellular processes and advancing personalized medicine. In evolutionary biology, my ongoing project models the extinction of asexual organisms, offering innovative alternatives to traditional explanations such as Muller’s Ratchet. Building on these works, I have been invited to contribute to multiple book chapters exploring mitochondrial cancer biology, with a particular focus on AI application in cancer research.

Democracy Measurement

I investigate democratic backsliding and institutional erosion by developing AI models to analyze political discourse, institutional performance metrics, and socio-economic indicators. This work provides early-warning systems for identifying potential democratic crises, offering invaluable tools to policymakers and international organizations. One notable aspect of this research is my effort to reduce biases in existing measures of democracy, such as those used by V-Dem. By employing computational methods, I address upward and downward biases in democratic backsliding data, providing a more accurate picture of institutional health. In an era where democratic institutions face unprecedented challenges, this work supports the development of robust frameworks for safeguarding democracy worldwide.

The Strike, Robert Koehler (1886)

The Anatomy Lesson of Dr Nicolaes Tulp, Rembrandt (1632)

The Ninth Wave, Ivan Aivazovsky, 1850

AI-generated by DALL·E, conceptualized with my assistant.

...There is only one thing you should do. Go into yourself. Find out the reason that commands you to write; see whether it has spread its roots into the very depths of your heart; confess to yourself whether you would have to die if you were forbidden to write. This most of all: ask yourself in the most silent hour of your night: must I write? Dig into yourself for a deep answer. And if this answer rings out in assent, if you meet this solemn question with a strong, simple “I must,” then build your life in accordance with this necessity; your while life, even into its humblest and most indifferent hour, must become a sign and witness to this impulse. Then come close to Nature...

Letters to a Young Poet, Rainer Maria Rilke 1875—1926