Hi, welcome to my website! :)

Research

My research sits at the intersection of NLP, clinical AI safety, and digital mental health. I am interested in building AI systems that are reliable and equitable in real-world clinical settings, with a particular focus on how language and passively collected digital signals can surface early warning signs in health contexts and support better clinical decision-making.

I do this by partnering with researchers across diverse fields, bringing NLP and AI evaluation methods into collaborative projects that tackle problems with real clinical stakes. My work ranges from how geographic distribution shift inflates errors in medical LLMs, to how AI conversations can be monitored continuously for mental health risk, to how multimodal signals can improve diagnostic accuracy in underserved clinical domains. I am also drawn to the gap between how these systems perform in controlled settings and how they behave when deployed on populations and in contexts they were never built to serve. Closing that gap, not just on benchmarks but in practice, is what drives the work.

My research sits at the intersection of NLP, clinical AI safety, and digital mental health. I am interested in building AI systems that are reliable and equitable in real-world clinical settings, with a particular focus on how language and passively collected digital signals can surface early warning signs in health contexts and support better clinical decision-making.

Out-of-Distribution

I am a massive Game of Thrones nerd. I prepare for every project like Winter is Coming, and when my models fail to converge, I simply whisper Valar Morghulis and try again. Moreover, I try to find any excuse to re-watch Interstellar. I have seen it too many times to count, but I still analyze the black hole scenes with the same focus I bring to my engineering work.