Postdoctorate in Machine Learning for Predictive Modeling of Mineral Fragmentation Behavior Bureau de Recherches Géologiques et Minières in France

Postdoctoral position in Machine Learning for Predictive Modeling of Mineral Fragmentation Behavior at Bureau de Recherches Géologiques et Minières (BRGM), France.

This postdoctoral role is part of the PEPR Sous-sol, bien commun / InnovTech project, specifically Sub-Task ST3.1.1, which aims to connect elementary ore fragmentation properties with the performance of comminution technologies. The research sits at the intersection of machine learning, geosciences, process engineering, and materials science, with a strong focus on sustainable mineral processing.

The postdoc will work on a challenging multimodal learning problem using three complementary data sources: 3D micro-CT volumes describing crack networks, porosity, and mineralogy; single-particle breakage test results including fracture energy and breakage probability; and DEM simulation outputs capturing fracture mechanisms and energy dissipation. The scientific goal is to learn a compact and transferable “fragmentability signature” of particle breakage behavior, then use it to predict fracture probability, fragment size distributions, loading sensitivity, energy consumption, and product size distribution at the process scale.

Research directions include multimodal data fusion, representation learning for complex material systems, probabilistic machine learning and uncertainty quantification, multi-scale predictive modeling, and data-efficient methods such as transfer learning, self-supervised learning, surrogate modeling, active learning, and uncertainty-aware decision-making. The position is especially suited to candidates interested in advanced ML methods applied to real scientific and industrial datasets.

The contract is full-time for 18 months, starting on 1 January 2027, within the DCGS/GDS unit at BRGM’s technical center in Orléans, France. The project involves close collaboration with a multidisciplinary team of researchers in data science, applied mathematics, process engineering, and materials science. The successful candidate will contribute to research publications, project deliverables, and presentations at international conferences.

Eligibility: a PhD in machine learning, AI, data science, applied mathematics, computational statistics, computer vision, scientific computing, or a closely related field. Strong ML/deep learning skills and experience with multimodal, probabilistic, or scientific ML approaches are highly desirable. Experience with Python and frameworks such as PyTorch or TensorFlow is expected. Mineral processing experience would be an advantage.

The application deadline is 27 November 2026. Applicants should submit a CV and cover letter via the BRGM recruitment portal. BRGM highlights a supportive work environment, teleworking possibilities, and employee benefits.