Postdoctoral Fellow

This position contributes to the HyperTails project — an interdisciplinary initiative led jointly by the research groups on Human Environmental Systems, Machine Learning and Complex Systems, and Remote Sensing. The project develops novel AI-based methods to detect and classify mine tailings ponds worldwide using hyperspectral satellite imagery and contextual information extracted from mining literature. The aim is to establish the first global dataset linking spectral reflectance to contaminant profiles, enabling scalable environmental risk screening of mine waste facilities. We are looking for a motivated person at the post-doc level who will be involved in the following tasks: Develop object-based detection of tailings facilities using machine-learning architectures (e.g. YOLO) and multispectral imagery (Sentinel-2). Process and analyse hyperspectral imagery (EnMAP) and spectral libraries to extract diagnostic features. Apply natural language processing to extract contextual information from mining-related grey literature. Integrate spectral and textual data to infer contaminant profiles and environmental risks, leveraging in-house expertise in inorganic chemistry and trace contaminants. Maintain Python-based workflows and GitHub repositories for the above tasks. The successful candidate holds a PhD in remote sensing, environmental science, geology, mineral exploration, or computer science, with proven expertise in in Python programming and version control, and either hyperspectral remote sensing or machine learning (ideally both). Experience in machine learning and deep learning (e.g. PyTorch, TensorFlow, scikit-learn), natural language processing or information extraction from unstructured text or hyperspectral remote sensing techniques (spectral angles, spectral unmixing, continuum removal) is an asset. Excellent English communication skills, analytical ability, and an independent working style are expected. Applications are reviewed on an on-going basis until the position is filled. Applications must include a motivation letter, a CV and two letters of recommendation. The period of appointment is 12 months with the option to extend by another 12 months, starting as agreed upon.