Phd position in mechanical engineering: ai-guided design of programmable intelligent metamaterials
Pubblicato il 17-08-2026 - The Italian Institute Of Artificial Intelligence in Roma
Ph D in Mechanical Engineering Research Title AI-Guided Design of Programmable Intelligent Metamaterials Deadline The 1st of July 2026 Funded by The Italian Institute of Artificial Intelligence (AI4 I), in collaboration with Politecnico di Milano Supervisors Marco Maurizi (AI4 I) Valentina Zega (Politecnico di Milano) Contacts (email hidden) (email hidden) Websites R& D Labs - AI4 I SCHEDA_5424_ISSG_AI-GUIDED_DESIGN_OF_PROGRAMMABLE. PDF Position Description The Ph D scholarship is funded by the Italian Institute of Artificial Intelligence (AI4 I). The research will be carried out jointly at AI4 I and Politecnico di Milano. The project focuses on architected materials, also known as metamaterials. The emergence of intelligent and re-programmable metamaterials — engineered systems whose functional properties arise from geometry and architecture rather than chemistry alone — marks a paradigm shift in how materials interact with their environment. Unlike conventional materials with fixed, intrinsic behavior, these metamaterials can sense external stimuli, process information,
and dynamically modify their response in real time. They embody a new class of programmable matter, where structural geometry and embedded functionality combine to achieve adaptive mechanical, acoustic, or electromechanical performance Such systems hold great promise in fields ranging from vibration and noise control to wearable haptics, soft robotics, and human–machine interfaces. However, the design and control of architectures capable of robust, reversible reconfiguration remain unsolved challenges. Research Methods and Techniques The research will combine machine learning, multi-physics simulation, and additive manufacturing in an integrated workflow for designing and realizing re-programmable metamaterials. Key methodologies could include: Generative deep learning models (Graph Neural Networks, Diffusion Models, Transformers) to explore design spaces and synthesize architectur