This fully funded PhD position at KU Leuven focuses on developing uncertainty-aware machine learning techniques for risk mitigation in safety-critical systems. Hosted at the Bruges Campus within the interdisciplinary M-Group, the research bridges computer science, electrical engineering, and mechatronics, with a strong emphasis on the dependability of intelligent systems. The successful candidate will join the Declarative Languages and Artificial Intelligence (DTAI) lab, renowned for its excellence in both fundamental and applied research in machine learning and artificial intelligence.

The project is part of the Flanders Make Strategic Basic Research project SAIfety, which aims to enhance the robustness and safety assurance of ML models throughout their lifecycle. The research will focus on uncertainty quantification methods, such as Bayesian and interval neural networks, to ensure reliable operation of ML-enabled safety-critical systems. Application domains include deep learning for computer vision and time-series analysis, with validation on industry-grade demonstrators. The work addresses real-world challenges, such as ensuring that ML models can signal when their predictions are unreliable, enabling systems to take appropriate safety measures in industrial environments.

KU Leuven offers a vibrant, internationally oriented research environment, ranked among the top 100 universities globally. The Bruges Campus provides access to state-of-the-art lab facilities and a dynamic team of AI researchers, with strong links to the mechatronics industry. Doctoral training is provided through the Arenberg Doctoral School, supporting both academic and professional development. The position includes opportunities to present research at international conferences, collaborate with industrial partners, and publish in high-impact venues.

Applicants should hold a Master's degree in Computer Science, Artificial Intelligence, Electrical Engineering, or a related field, with above-average academic performance. Experience with machine/deep learning, Python, and data science toolkits is required; prior exposure to uncertainty quantification, computer vision, or time series analysis is advantageous. Candidates must demonstrate strong communication skills, independence, and a collaborative spirit. KU Leuven values diversity and encourages candidates from all backgrounds to apply.

The position is fully funded for four years, with a competitive remuneration package, health insurance, and university benefits. The application deadline is January 22, 2026. To apply, use the online application tool and submit all required documents in a single PDF. For further information, contact Prof. Mathias Verbeke or Dr. Laurens Devos by email, mentioning [UQ Vacancy] in the subject line.