This postdoctoral position at the University of Montpellier is designed for researchers interested in developing advanced artificial intelligence models to predict the abundance and distribution of fish, plankton, and organic matter in marine environments. The successful candidate will join the MARine Biodiversity Exploitation & Conservation (MARBEC) laboratory, a leading research unit focused on marine biodiversity, with additional workspace and resources at the LIRMM laboratory. The project is supported by several EU initiatives, including the Horizon Europe programme, and provides access to state-of-the-art computational resources such as the Jean Zay supercomputer and shared GPU facilities. The main mission is to create predictive models that incorporate size classes, taxa, morphotypes, and spatiotemporal anthropogenic and socioeconomic factors. Key activities include integrating seascape-level environmental and human factors using deep learning and vision transformer models, leveraging transfer learning and foundation models like Prithvi-EO-2.0, utilizing large language models for ecological effect prediction, and applying explainable AI to quantify impacts on biodiversity. The position offers an annual budget, personal computer, and support for participation in scientific congresses. The work environment is highly collaborative and interdisciplinary, providing opportunities to work with leading researchers and access cutting-edge technology. Applicants should have a PhD in ecological, marine, or computer science, with strong skills in Python and/or R, QGIS, database management, and modeling. Experience in deep learning, vision transformers, and predictive modeling is highly desirable. The application deadline is November 21, 2025. Interested candidates should send their application materials to Dr. David Mouillot at david.mouillot@umontpellier.fr, including a CV, cover letter, and relevant documentation. For further details, candidates can visit the MARBEC laboratory website or the Euraxess position link.