← Toutes les offres
EZ

Research Scientist in Scientific AI and Geotechnical Engineering

ETH Zürich

Zürich, Zh
CDD
Publié le 13 novembre 2026 · Expire le 13 décembre 2026

Description du poste

Einleitungstext The Chair of Geotechnical Engineering at ETH Zurich, part of the Institute for Geotechnical Engineering (IGT), focuses its teaching and research on foundation engineering, soil mechanics and natural hazards. Our objective is to combine theoretical knowledge with experiments and numerical simulations to develop new methods for understanding and predicting the behaviour of soils, foundations and infrastructure. We are opening a position in our research group at ETH Zurich, focusing on developing and integrating modern Artificial Intelligence (AI) methods into geotechnical engineering research, while maintaining a strong connection to the underlying physics and mechanics. We do not aim to develop a new AI model or replace physics-based engineering with purely data-driven AI. Instead, we want to use AI to transform the way research itself is performed—from numerical model generation and experimental data acquisition to analysis, visualization and communication. For this purpose, we are looking for a Research Scientist in Scientific AI and Geotechnical Engineering, who will work across the research group, collaborating with PhD students, postdoctoral researchers and senior scientists. The role combines scientific research, computational engineering, research automation, data science and the development of reusable AI-enabled tools. A central objective is to identify new opportunities where AI can enable research that would otherwise be difficult or impossible, and to develop capabilities that can be used across the entire group. Aufgaben-/Stellenbeschreibung The job will interconnect several different areas of our activity: Scientific AI and physics-informed modelling Develop and apply AI and machine-learning approaches to challenging problems in geotechnical engineering and computational mechanics, with particular emphasis on approaches that combine data with physical knowledge, rather than relying exclusively on black-box, data-driven models. Explore to