I'm a Machine Learning Research Scientist at NuMind, where I work on supervised fine-tuning and reinforcement learning for LLMs specialized in structured information extraction.
My background spans explainable AI, computer vision, and geospatial AI: I hold a PhD on deep learning for power systems (RTE / Mines Paris), and an MSc in Machine Learning from ENSAE.
I'm also the maintainer of DeepPVMapper, an open-source, collaborative project for large-scale mapping of rooftop photovoltaic systems.
Outside of work, you'll usually find me out on a run.
Latest blog posts
- Rooftop Solar Needs an Open Registry Why DeepPVMapper and PV Check exist, and where it is going
- If in Doubt, Say So A crowd of volunteers, a detector, and the value of being allowed to say “not sure”
- Navier–Stokes, Scaling Laws, and Peak Tokens What if the real constraint on AI isn't energy or money, but tokens — and the race to the frontier is built to cannibalize itself?