I am a DPhil student at the University of Oxford and a co-founder of the Oxford LLMs initiative, a research and education initiative focused on the intersection of large language models and social science.

My research examines the extent to which LLMs can replicate human survey responses and predict public opinion,  probing where models succeed, where they fail, and how their capabilities for this task can be rigorously evaluated. This work draws on cross-national public opinion data and engages directly with questions of measurement validity and model reliability.

Alongside my research, I am an active educator. I teach the LLMs for Social Science course across multiple European institutions — Oxford, Paris, and Florence — and have designed workshops for social science PhD students on integrating LLMs into empirical research workflows. I also serve as a teaching assistant for introductory machine learning courses at Oxford.