Postdoc in Scientific Machine Learning and Interpretable Physics-AI for Geosystems at Geological Survey of Finland Geological Survey of Finland in Finland

The Geological Survey of Finland (GTK) is offering a postdoctoral position in Scientific Machine Learning and Interpretable Physics-AI for Geosystems. This research opportunity is situated within the Water and Mining Environment Solutions unit, focusing on the development of efficient and reliable surrogate models for highly nonlinear PDEs describing coupled groundwater contaminant transport dynamics and mitigation in subsurface systems. The project integrates advanced scientific machine learning (SciML) methodologies, including neural operators and physics-informed neural networks (PINNs), with process-based reactive transport modeling. The research aims to advance hybrid modeling frameworks at the intersection of AI, physics, and geoscience, providing strong international visibility and collaboration opportunities.

The successful candidate will join a multinational team of around 50 experts working on groundwater research, hydrogeochemistry, groundwater flow, reactive transport modeling, mining environmental issues, and water management. The group is multi-locational and offers flexible remote work options within Finland. The project is supported by the Research Council of Finland and includes collaboration with the University of Eastern Finland and an active international network, offering opportunities for research visits abroad and extended scientific exchange with leading research groups.

Applicants must hold a PhD in applied mathematics, computational science, computer science, hydrogeology, or a related field before the start date. Essential qualifications include a background in scientific machine learning, neural operators, or physics-informed learning, a strong publication record, experience with deep learning libraries, PDE-based modeling, scientific computing, and numerical methods, as well as fluency in English. Additional valued skills are experience in reactive transport modeling, inverse problems, data assimilation, uncertainty-aware modeling, and numerical methods related to environmental or subsurface modeling. Candidates who have defended and are awaiting their diploma or are expected to defend before the start date are also eligible.

The position is fully funded until the end of 2028, with a starting salary of EUR 4414-4704 per month, central government holiday benefits, flexible working hours, and the possibility to work remotely in Finland. The fixed-term contract runs from 1 May 2026 to 31 December 2028. Applications must be submitted in English via the valtiolle.fi service by 25 February 2026 at 16:15 (EET), including a motivation letter and CV. For further information, contact Research Scientist Hanli Qiao ([email protected]) or Team Manager Anna Tornivaara ([email protected]).

Keywords: scientific machine learning, interpretable physics-AI, geosystems, reactive transport modeling, groundwater contaminant transport, neural operators, physics-informed neural networks, hydrogeology, deep learning, environmental modeling.