PhD in Hardware Architectures for Physics-Informed AI

This PhD position at Eindhoven University of Technology (TU/e) focuses on designing novel hardware architectures for physics-informed AI models, with a strong emphasis on FPGA and ASIC-based acceleration. The project aims to bridge the gap between computationally demanding AI methods and efficient hardware deployment, enabling real-time operation in scientific, industrial, and sensing applications. You will collaborate with the Electrical Engineering department and contribute to the NWO OTP project on counterfeit cash detection using Physics-Infused Deep Unfolding (PIDU), while also developing broader design methodologies for physics-informed AI techniques. The role involves algorithm development, hardware-software co-design, and practical demonstrator creation, with opportunities for interdisciplinary collaboration and teaching.

• Investigate novel hardware architectures for physics-informed AI models, focusing on FPGA/ASIC-based acceleration and edge deployment.
• Develop hybrid model-based and learning-based algorithms optimized for reconfigurable hardware platforms.
• Design and implement real-time demonstrators, including a counterfeit cash detection system using Physics-Infused Deep Unfolding (PIDU).
• Collaborate with industrial partners (e.g., Smiths Detection, Sioux Technologies) to integrate research findings into practical applications.
• Contribute to both fundamental research and applied demonstrators within the NWO OTP project framework.
• Develop teaching skills and mentor students as part of the PhD program.
• Publish research findings in top-tier conferences and journals within the field.