We invite applications for a project scientist (post-doc) position at the GeoHydrodynamics and Environment Research group (GHER, https://www.gher.uliege.be) of the University of Liège in Belgium. The candidate will work as a member of the GHER contributing to the EU research project COMEDI in a consortium of 11 leading partners in the field of data assimilation and deep learning. A successful applicant will develop and adapt assimilation schemes based on generative deep learning methods (such as flow matching and diffusion models).
The candidate should have previous experience in data assimilation and/or deep learning as well as a strong background in scientific programming (in languages like Julia, Python, Fortran or C/C++). The applicant must hold a PhD in physical oceanography, atmospheric sciences, computer vision, engineering or related disciplines. Proficiency in English communication skills (oral and written) is expected for international collaboration.
This position falls under a program promoting scientific mobility. Researchers must not have resided or carried out their main activity (employment, studies, etc.) in Belgium for more than 24 months during the three years immediately preceding the start date of the contract. Short stays (such as vacation and attendance to conferences) are not taken into account.
The duration of the contract will be 3 years. The tentative starting date of the position is the 1st January 2027. The job posting is also available at GHER - Jobs. The page at the previous link will be updated once the position is filled.
Please send the application dossier to Alexander Barth (a.barth at uliege dot be). The applications will be continuously reviewed until the position is filled. Short-listed applications will be contacted via email. The dossier should contain:
- A full CV including with a list of peer-reviewed publication and corresponding DOIs
- The full PDF of the most significant peer-reviewed publication (chosen by the applicant).
- Motivation letter (expressed in their own words, no Large Language Models output please, maximum 1 page)