Freelancer profile translated to English.

Description

I build mathematical models and physical simulations to solve problems that classical approaches cannot handle — because I understand both physics and code.

With 15+ years of experience in numerical simulation, mathematical modeling, and machine learning, I work at the intersection of fundamental science and engineering problems. My recent work includes developing physics-informed neural networks (PINNs) in PyTorch for inverse material design (École Polytechnique / CNRS), C++ simulation frameworks for 3D surface mechanics, and predictive models for complex mechanical networks published in Physical Review Research and Physical Review Letters.

I take on assignments where the problem is genuinely difficult: systems governed by PDEs, inverse problems, mechanical simulation, geometric modeling (DDG, NURBS), or ML applied to physical systems where purely data-driven approaches fail.

I work autonomously, deliver clear results, and can communicate with technical and non-technical audiences.

Available immediately for scientific computing, simulation, and applied ML assignments.

Languages: English (native), French (B2),

Industry field of expertise

Languages

English
Native or bilingual
French
Fluent

Workplace preferences

Can work on-site

Lyon (up to 50km)

CNRS / IGFL / ENS-Lyon & École Polytechnique,
Researcher – 3D Geometry, ML & Computational Physics
BIOTECH
October 2024 - March 2026 (1 year and 5 months)
France
- Design and training of PINNs in PyTorch for 3D material design (MATERIALIZE Project, École Polytechnique)
- C++ simulation framework for 3D surface mechanics based on discrete differential geometry on meshes (ERC project)
- NURBS representations for modeling growing biological surfaces
- Short-cycle R&D projects with clear deliverables; multidisciplinary collaboration
Python Physics Physics-informed Neural Networks C++ Modeling
École Polytechnique,
Postdoctoral Researcher – Computational Geometry
MECHANICAL ENGINEERING
October 2021 - September 2024 (2 years and 11 months)
Laboratoire d'Hydrodynamique de l'École polytechnique (LadHyX), Palaiseau, France
- Predictive numerical methods (Python) for 3D morphogenesis and material response
- Predictive models for the mechanical properties of complex geometric networks (published in Physical Review Research)
- Co-supervision of M.Sc. student: measuring 3D surface flatness using computational geometry
Python Modeling Physics Physics-Informed Neural Networks (PINN) Data science
Collège de France,
Postdoctoral Researcher – Analytical Geometry & Mechanics
RESEARCH
October 2019 - October 2021 (2 years)
Paris, France
- Exact analytical models for 3D tissue geometry, instabilities, and rheology (published in Physical Review Letters)
Modeling Applied Mathematics Physics differential geometry statistical physics