My work sits at the intersection of international relations and AI systems.
A lot of my thinking is shaped by game theory, particularly in understanding AI races: how competition shifts behavior, where coordination breaks down, and how those dynamics produce downstream consequences for countries with limited bargaining power.
On the technical side, I focus on AI evaluation and interpretability, exploring how models reason, where they fail, and how we might assess them more rigorously.
These threads meet in technical governance, the messy space where models, institutions, and incentives collide through decisions around training, deployment, and evaluation.
This site is my digital garden, project archive, and research notebook. Some pages are polished essays, some are project writeups, and some are still fragments. I want it to feel alive rather than finished.