The Data Science for Sciences team researches at the frontier of machine learning, computer vision, and remote sensing to solve scientific questions in the environmental and earth sciences. Our goal is to develop original, data-driven methods to automatically analyse environmental data at very large scales. We work closely with our colleagues in ecology to jointly find new ways to protect our environment on a global scale using artificial intelligence. Scientific projects include global mapping of vegetation parameters such as tree canopy height and carbon stock with very high spatial and temporal resolution, monitoring agricultural land, predicting water levels in flooding scenarios or setting up a rapid alert system to detect forest degradation. On the technical side, we are investigating exciting topics such as quantifying uncertainties in deep learning, explainable artificial intelligence, graphical neural networks, and time series analysis with transformers. We believe that interdisciplinary research is the key to scientific breakthroughs and we always strive to put our research into practice by collaborating with NGOs, companies, and public administration.