Postdoctoral Fellowship in Machine Learning for Turbulent Dynamo Modeling at Université Paris-Saclay

This two-year postdoctoral fellowship at Université Paris-Saclay focuses on machine learning for turbulent dynamo modeling within the MilaDy ANR project. The successful candidate will join the LISN CNRS laboratory and work on large-scale numerical simulations of magnetic and velocity fields, investigating magnetic field generation via the dynamo effect. The research involves close collaboration with the Lagrange Laboratory (Nice) and CEREA (ENPC, EDF R&D). The project centers on magnetohydrodynamics (MHD), specifically the analysis of magnetic field reversals and dynamo burst phenomena using reduced-order models (ROMs) and advanced machine learning techniques. The scientific objectives include developing parametric reduced-order models with enhanced physical interpretability, optimizing experiments using these models, and overcoming current limitations in model generalization and extrapolation. The research will leverage physics-informed neural networks, variational autoencoders, and other deep learning methods to ensure robust and explainable models. Candidates should have a PhD in fluid mechanics, applied mathematics, or machine learning, with strong multidisciplinary skills in fluid dynamics, MHD, scientific data analysis, and programming. The position offers a gross monthly salary between €3,131.32 and €3,569.85, depending on experience, and is based at LISN, Université Paris-Saclay, Orsay. The research environment is highly interdisciplinary, with access to high-performance computing resources and advanced simulation databases. To apply, candidates should send a single PDF containing their CV, motivation letter, list of papers, and recommendation letters to the provided contact emails, using the specified subject line. Applications are accepted in English or French. The position is subject to authorization from the French Ministry of Higher Education and Research.