Olga Fink

EPFL’s IMOS Lab is advertising a PhD position in Physics-Informed Graph Neural Networks for Wind Turbine Health Monitoring . The project sits at the intersection of computer science , physics , and mechanical engineering , with a strong focus on data-driven methods for predictive maintenance and structural/asset health monitoring in renewable energy systems. The position is associated with Olga Fink , Assistant Professor of Intelligent Maintenance and Operations Systems at EPFL. The research topic suggests work on graph neural networks, physics-informed machine learning, and condition monitoring for wind turbines, likely involving sensor data, diagnostics, and reliability-oriented modeling. This is a PhD opening at École polytechnique fédérale de Lausanne (EPFL) in Lausanne, Switzerland. The post does not mention a deadline, stipend amount, or detailed eligibility criteria in the excerpt provided. It does indicate that applicants should be highly motivated. Interested candidates should review the official EPFL careers page and apply through the job portal linked in the post. The LinkedIn post also points to the same opportunity for additional context.

Postdoctoral Researcher in Tabular Foundation Models for Building and District Energy Systems This postdoctoral opportunity is a collaboration between the Urban Energy Systems Laboratory at EMPA and the Intelligent Maintenance and Operations Systems (IMOS) Laboratory at École Polytechnique Fédérale de Lausanne (EPFL) . The project sits at the intersection of computer science , electrical engineering , mechanical engineering , physics , and statistics , with a strong focus on machine learning for engineering systems. The research aims to advance tabular foundation models that can transfer across heterogeneous energy systems and tasks with limited task-specific data. Key topics include efficient adaptation of pretrained models, incorporating physical knowledge into foundation-model architectures and learning objectives, and generalization across datasets with different features, operating conditions, and target quantities. The testbed is building and district energy systems , with applications in prediction, surrogate modeling, uncertainty quantification, and decision support. The post is especially suitable for candidates with a PhD and a strong publication or research record in machine learning, particularly in foundation-model architectures, pretraining, and efficient adaptation. Experience with tabular foundation models or heterogeneous structured data is highly relevant. Interest in engineering applications is essential, and prior exposure to energy systems is welcome. The position starts immediately or by agreement. Applicants are asked to apply via the online portal and include a motivation letter, CV, a one-page research proposal, certificates and diplomas, and contact details for two referees. Funding details are not specified in the post. The opportunity is based in Switzerland.