Matthias Gerstgrasser’s personal website

I am a research scientist at OpenAI. Previously I was a postdoc at Stanford Computer Science working with Diyi Yang and at Harvard University’s School of Engineering and Applied Sciences working with David Parkes. I am particularly interested in multi-agent learning and reasoning. In lay terms, I teach AIs to play nice with each other, and with us humans.

I did my DPhil (PhD) at Oxford with Paul Goldberg and Elias Koutsoupias. My thesis was on mechanism design in two-sided markets, a step toward embedding mechanism design in larger economies.

Beyond academic settings, I have held industry research positions at institutions such as the European Space Agency’s Advanced Concepts Team, IBM Research Kenya, and ARM’s System Design Division. I care about translating research into applications and have an interest in entrepreneurship. I have participated in the Entrepreneur First incubator programme, developed a “lean”-type business during my PhD, and more recently co-founded an electronics startup, now post-revenue.

I am head of ASciNA Bay Area and previously served as head of ASciNA Boston and president of the Oxford University Austrian Society.