Ricardo Torres

Wageningen University & Research is advertising a fully funded PhD position in Foundation Models for Agricultural Sciences within the Artificial Intelligence group. The project sits in the AgriscienceFM initiative and focuses on developing and evaluating domain-specific AI foundation models for agriculture, especially where standard models struggle to generalize across real-world agricultural settings. The research combines computer science , artificial intelligence , machine learning , agricultural sciences , and environmental science . Topics include self-supervised learning , contrastive learning , physics-informed and knowledge-guided ML , remote sensing , climate data , earth observation , time-series analysis , crop type classification, yield forecasting, field boundary delineation, crop disease detection, and crop failure detection. The work involves multi-modal heterogeneous data such as text, images, location data, and time series, and large-scale training on HPC systems. The PhD is embedded in an interdisciplinary and international team led by Prof. Ioannis Athanasiadis , with co-supervision by Prof. Ricardo Torres and Dr. Taniya Kapoor . The position is based in Wageningen, Netherlands , at one of the world’s leading life sciences universities. Eligibility highlights include an MSc in AI, Computer Science, Engineering, or a related field , demonstrated experience in applied machine learning, preferably in remote sensing or agriculture, strong Python skills, and familiarity with PyTorch , Scikit-Learn , or similar tools. Strong writing skills are required, and English proficiency at C1 level is expected. Funding includes a fully funded PhD salary of €3,059 to €3,881 per month over 4 years, plus a tailored training program, pension, year-end bonus, sports facilities, and visa/relocation support. The initial contract is for 18 months and may be extended to the full project duration. Application deadline: 5 May 2026. Applicants should submit a CV, motivation letter, and one scientific writing sample, each limited to 3 pages. Transcripts are not required at this stage, and applications must be submitted via the official WUR vacancy page.