About me

I am a third-year Ph.D. student in the WILLOW team at Inria Paris and the Computer Science Department of École Normale Supérieure (ENS). I am advised by Justin Carpentier (Inria & ENS Paris) and Nicolas Perrin-Gilbert (ISIR, Sorbonne University).

My research focuses on improving robots’ agility and dexterity through the integration of optimization and reinforcement learning. I explore sampling-based optimization methods enhanced with gradient information and expressive policy representations such as flow-matching. My goal is to design efficient algorithms for control and optimization, covering both theoretical foundations and real-world deployement.

Previously, I was a Research Engineer at Wandercraft, where I applied RL to control a self-balancing exoskeleton. I also served as a core developer for open-source libraries at Inria, contributing to Pinocchio and Proxsuite. I hold a master’s degree in Robotics, Cognition and Intelligence from Technical University Munich (TUM), obtained with distinction in January 2021. During my studies, I conducted research on deep learning for robotics projects in Germany and Japan under the supervision of Prof. Laura Leal-Taixé and Prof. Yoshihiko Nakamura.