Welcome on my webpage ! I am a temporary research and teaching fellow in mathematics at ENS Lyon and part of the 'Probability' team of UMPA. Before this, I have completed my PhD at Institut Fourier, under the supervision of Vincent Beffara.
My main research interest is about statistical mechanics and percolation.
At the same time, I teach preparatory courses for the agrégation to master's students.

Detailed CV.

PhD Thesis: Glauber dynamics, factors of product measures and Gibbs measures, defended on October 16th, 2025 (Thesis) (Slides, in French)

Research

In statistical physics, we aim to understand the macroscopic behaviour of physical systems on the basis of microscopic interactions between particles. A particularly interesting phenomenon is the phase transition, which consists in an abrupt change in the behaviour of the system around a certain critical parameter (e.g. temperature). In 'real life', these include solid-liquid and liquid-gas transitions, as well as the so-called Curie point, which is the temperature at which a ferromagnetic material loses its magnetisation. Mathematically, this last phenomenon can be interpreted in the formalism of the Ising model, which is probably the reference model in statistical physics presenting a phase transition (in dimension d ≥ 2).
If the phase diagram of the Ising model is very well known today, it is mostly thanks to the existence of correlation inequalities of the 'FKG' type. In my PhD thesis, I am interested in non-FKG methods that can be applied to models that are not positively correlated, as is the case for Ising at negative temperature, the random-cluster model with q<1, or even disordered systems with the addition of random coupling constants, used to model spin glasses.

Publications and preprints

• "A new bound for the critical point of the FK model for q<1" (2025). arXiv.
• "Glauber dynamics and coupling-from-the-past for Gaussian fields", Electron. J. Probab. 31: 1-38 (2026). Journal / arXiv.
• "Subcritical Boolean percolation on graphs of bounded degree", Electron. Commun. Probab. 30: 1-12, (2025). Journal / arXiv.

Teaching

This year, 2025-2026, I am teaching preparatory classes for the agrégation at ENS Lyon, mainly probability and statistics courses.
Between 2021 and 2025, I taught several tutorials and courses/tutorials at UGA, in the first and second years of the mathematics, computer science or physics degree programme.

Talks

• Mar. 2026 : Les probas du vendredi, LPSM, Paris (Abstract)
• Feb. 2026 : Les probabilités de demain, Institut Henri Poincaré, Paris (Program of the conference)
• Feb. 2026 : Workshop in Probability, Ergodic Theory and Dynamical Systems, Rouen (Abstract)
• Jan. 2026 : Probability seminar, ICJ/UMPA, Lyon (Abstract)
• Jan. 2026 : Probability seminar, I2M, Marseille (Abstract)
• Nov. 2025 : Oberseminar Stochastik, Institute for Applied Mathematics, Bonn University (Abstract)
• Sep. 2025 : Probability seminar, Institut Fourier, Grenoble (Abstract)
• April 2025 : SPACE Seminar, Institut Denis Poisson, Tours (Abstract)
• Jan. 2025 : Drafting workshop in Discrete Mathematics and Probability, Rényi Institute, Budapest (Webpage of the workshop)
• July 2024 : Probability summer school, Saint-Flour (Webpage the summer school)
• Nov. 2023 : "Comprehensible" PhD students seminar, Institut Fourier, Grenoble (Abstract)
• Oct. 2023 : Colloquium Young Probabilists and Statisticians, Saint-Pierre d'Oléron (Webpage of the colloquium)
• June 2023 : PPPP 2023, Superconcentration and chaos, Grenoble (Webpage of the conference)
• April 2023 : Probability seminar, Institut Fourier, Grenoble (Abstract)
• Oct. 2022 : PhD days, Institut Fourier, Grenoble (Program)

I co-organised a working group at the Institut Fourier, in which every week (or so!) a PhD student from the probability team or an M2 intern presents a research article recently posted on arXiv, on various topics in probability. You can find the list of talks here.
On Philippe Caldero's youtube channel, you'll find a series of videos I made (in French) on the sensitivity conjecture, proved by Hao Huang in 2019, on the complexity of Boolean functions.

Misc.

Here is a manifesto for limiting the use of planes in maths labs (in French), initiated by colleagues in Lyon and Grenoble.