Postdoctoral Position in Machine Learning for Automated Plant Phenotyping (PhenoMix Project) ETH Zürich in Switzerland

The Swiss Data Science Center (SDSC) at ETH Zürich is offering a postdoctoral position as part of the PhenoMix project, funded by the Swiss National Science Foundation (SNSF). This interdisciplinary project focuses on developing advanced machine learning and computer vision methods for automated plant phenotyping, with a strong emphasis on sustainable agriculture and crop science. The successful candidate will join a collaborative team spanning SDSC, the Crop Science Group (Prof. Achim Walter), the Grassland Sciences Group (Prof. Nina Buchmann), and AGROSCOPE (Dr. Susanne Vogelgsang), working at the intersection of data science, agricultural sciences, and environmental systems.

The PhenoMix project leverages the Field Imaging Platform (FIP), a high-throughput phenotyping facility, and field experiments to generate multi-modal datasets of pure stands and crop mixtures. The postdoctoral researcher will develop novel data science tools for automated trait estimation, including foundation models for phenotyping, domain transfer methods, 3D reconstruction and rendering, human-in-the-loop approaches, and rigorous field evaluation. The role involves creating models that generalize across imaging platforms and environmental conditions, with real-world impact for farmers, breeders, and researchers.

Key responsibilities include designing and implementing machine learning approaches for multi-trait plant phenotyping, developing domain-specific and physiologically plausible models, deploying active learning strategies, conducting field experiments, and generating comprehensive datasets for downstream analyses. The postdoc will also contribute to codebases, engage with open source communities, supervise students, and prepare scientific publications for top-tier venues.

Applicants must have a PhD in computer science, machine learning, data science, or a relevant domain science, with demonstrated expertise in machine learning and computer vision. Required skills include proficiency in deep learning frameworks (PyTorch preferred), scientific programming in Python, experience with large multi-modal datasets, and excellent communication skills in English. Beneficial competencies include 3D reconstruction, active learning, Bayesian optimisation, and familiarity with agricultural sciences.

The position is fully funded for up to 4 years, offering access to state-of-the-art phenotyping infrastructure, computational resources, and opportunities for professional development, including publishing research, presenting at international events, and supervising MSc and BSc students. ETH Zürich values diversity, sustainability, and work-life balance, providing a stimulating and inclusive research environment in Zurich.

Applications must be submitted online via the ETH Zurich application portal. Required documents include a letter of motivation, CV with publication list, academic diplomas, transcripts, certificates, and contact details for 2–3 references. Links to code repositories or portfolios may be included. For questions regarding the position, contact Dr. Michele Volpi ([email protected]). Applications via email or postal services will not be considered.

ETH Zürich is a world-leading university specializing in science and technology, renowned for its excellent education, cutting-edge research, and commitment to solving global challenges. The PhenoMix project offers a unique opportunity to contribute to the advancement of automated plant phenotyping and sustainable agriculture through innovative machine learning research.