Matthieu Scharffe
Engineer · AI & Robotics · Building
Engineer · AI & Robotics · Building
Fresh graduate from CentraleSupélec and EPFL with a Master's in Mechanical Engineering. Did cutting-edge AI research at 👑 Rolex and at VITA Lab at EPFL. I'm now working at the intersection of control theory, robotics, and machine learning - from deriving and simulating physical models to training reinforcement-learning policies and shipping the code that runs them.
Shipping products, learning fast
Reinforcement-learning agent that plays the real Temple Run on a physical phone — via a Frida gadget and a reverse-engineered simulator. Chasing a reliable 10,000 m.
I created a custom algorithm that tracks progress and generates personalized weekly meal plans
AI agent to help build CGI animations shorts
Master's thesis at EPFL VITA Lab. Built a benchmark of 1,500+ synthetic driving videos and a training-free tool-augmented agent that triples Qwen3-VL accuracy on edge-case driving judgment (21.9% → 60.1%). Paper in preparation.
Modeled the mechanical behavior of a flapping wing using PDEs from continuum mechanics. Solved using finite differences in Python and conducted DIC experiments.
Led a team of 6 students for two years. Built an autonomous hexacopter and a 1.8m wingspan VTOL to perform onboard person recognition missions using state-of-the-art technology.
Master of Mechanical Engineering - Lausanne, Switzerland
Engineering school - Paris, France