Antoine Levitt

Antoine Levitt is an Assistant Professor (CPJ) at the Laboratoire de Mathématiques d'Orsay, Université Paris-Saclay. Previously, he was a researcher at Inria Paris in the MATHERIALS project-team, affiliated with CERMICS at École des Ponts. His research focuses on numerical methods for quantum chemistry and materials science, including density functional theory (DFT), Wannier functions, and electronic structure calculations. He has contributed to theoretical advances in numerical analysis, such as error bounds for DFT and preconditioning techniques for eigenvalue problems.

Levitt’s work spans computational physics and applied mathematics, with applications to quantum dynamics, response properties, and defect systems. He has developed open-source software tools like DFTK.jl and Wannierization algorithms. His teaching includes courses on scientific computing, functional analysis, and quantum mechanics at both undergraduate and graduate levels.

Key research themes include numerical stability in DFT, analysis of linear response theory, and efficient implementations of conjugate gradient methods. His recent work explores machine learning applications in canonical sampling and mathematical foundations of Density Matrix Embedding Theory.