I am a postdoctoral researcher in the Earth Observation of Agroecosystems team at Agroscope, working at the intersection of artificial intelligence and Earth observation. I completed my Ph.D. in the EcoVision Lab at ETH Zurich, advised by Prof. Konrad Schindler and Prof. Jan Wegner. My research interests include foundation models, time-series and sequence modeling, uncertainty quantification, and agentic systems, particularly for Earth observation. My broader interests span fine-grained image classification, visual camera localization, and generative methods for visual synthesis.

I hold bachelor’s degrees in Physics and Electrical Engineering from Boğaziçi University and a master’s degree in Electrical Engineering, specializing in computer vision, from the University of Twente. For my master’s thesis, I investigated sequential image generation using generative adversarial networks at the University of Amsterdam’s Deep Vision Lab.

My experience spans academic research and industry. During my studies, I worked on visual camera localization at Niantic’s AR Lab in London, 3D object recognition for mobile robots at TNO’s Intelligent Imaging Lab in The Hague, and robotics software engineering at Robotnik in Valencia. Before starting my Ph.D., I worked as a 3D computer graphics R&D engineer at Colormass in Berlin.

After my Ph.D., I co-founded ModaMind, an AI fashion startup that generates realistic imagery for virtual fashion photoshoots. I also serve as a scientific advisor at Globeholder AI, which focuses on agentic Type-2 planetary reasoning.