About me

I’m a research scientist and medical doctor working on large language models for high-stakes settings, with a focus on alignment, evaluation, and safe deployment in clinical care. I recently defended my PhD in Computer Science at the Université Catholique de Louvain, where I led the development of Internist-7B — the first 7B medical LLM to surpass 60% on MedQA — and built an Epic-integrated clinical agent now used by thousands of clinicians at the Cliniques Universitaires Saint-Luc. From April 2026, I’m continuing this work as a postdoctoral researcher at Stanford, designing medical benchmarks and red-teaming multimodal frontier models. My broader interest is in trustworthy reasoning, scalable oversight, and the engineering needed to make frontier models reliable in real-world environments.