Are you passionate about combining artificial intelligence, physics-based modelling, and digital hardware design? Do you want to develop the next generation of FPGA/ASIC-based computing platforms that make physics-informed AI fast, efficient, and deployable in real-world mission-critical sensing systems? Physics-informed AI is emerging as a powerful alternative to purely data-driven machine learning. By incorporating physical models, domain knowledge, and optimization algorithms directly into learning systems, physics-informed AI can achieve higher accuracy, better generalization, greater interpretability, and significantly reduced training-data requirements. Despite these advantages, many physics-informed AI methods remain computationally demanding and are often developed without considering efficient hardware implementation. In this PhD project, you will investigate novel hardware architectures for physics-informed AI models, focusing on FPGA/ASIC-based acceleration and edge deployment. The research will explore how hybrid model-based and learning-based algorithms can be mapped efficiently onto reconfigurable hardware platforms, enabling real-time operation in scientific, industrial, and sensing applications. You will be part of a multidisciplinary collaboration with the EE department and contribute to the NWO OTP project "Detection of Hidden Cash using Physics-Based AI" through algorithm development, hardware architecture design, and hardware-software co-design. You will also have opportunities to contribute to both fundamental research and practical demonstrators. One of the project goals is to build a real-time demonstrator of counterfeit cash detection using Physics-Infused Deep Unfolding (PIDU) in collaboration with project stakeholders like Smiths Detection, Sioux Technologies, and the Dutch Customs. While PIDU will serve as an important research vehicle, the PhD will not be limited to this framework. The broader objective is to develop design methodologies and hardware architectures applicable to a wide range of physics-informed AI techniques, including deep unfolding, model-based learning, hybrid optimization-learning methods, and physics-informed neural networks.