I lead the NLP research at MICS in 🎓 CentraleSupélec (Paris-Saclay) and ⚖️ Equall.ai. My work focuses on making large language models 🤖 reliable, 📏 measurable, and 🚀 deployable in high-stakes domains where precision matters—like law, healthcare, and enterprise systems.
🏛️ Equall.ai — AI for Law
Legal AI demands precision and domain expertise. We built SaulLM—the first LLM for law—and use it to power M&A due diligence automation currently used at Big Laws.
I believe AI research should be open, reproducible, and useful. I build models that anyone can use, develop evaluation methods that reveal true capabilities, and focus on problems that matter in the real world. Good research creates tools that others can build on—not just papers.