Hi there! I'm Jo. As of fall 2025, I study mathematics and philosophy at UChicago and do research in AI Safety. I also co-started the UChicago AI Safety Group at XLab.
I am interested in learning and generalization dynamics of machine learning models, especially as it pertains to (mis)alignment. Currently, I'm working to better understand how reinforcement learning on narrow, real-world objectives shift model values, fortunately mentored by James Evans and Austin Kozlowski at UChicago Knowledge Lab.
I want to be the best version of myself possible. If you see a way I can improve (in how I think, how I treat people, how I engage with the world), let me know here. I promise to 1) take your feedback seriously, 2) not guess who you are, 3) assume good faith.
I enjoy thinking about thinking. My most recent experiment is using spaced repetition systems to help compress and schematize mathematical knowledge. If you have thoughts on this, I would love to compare notes.
The thinkers, writers, and artists who have most shaped me are Immanuel Kant, Alice Munro, Jorge Luis Borges, and Judit Reigl. For me, they represent structure, restraint, imagination, and rhythm.
Meditation is a central part of my life. I practice two lineages: Vipassana in the tradition of Goenka, and Taoism, from my time at the Five Immortals Temple.