DCs

Yunfan Zhang

Originally from China, Yunfan Zhang completed his Diplom’s degree in 2023 in Mechanical Engineering at TU Dresden. He worked as a research assistant at Fraunhofer IKTS and HZDR, developing modern deep learning and computer vision algorithms for various industrial scenarios including robotic inkjet printing, battery recycling, and X-ray holography inspection. His Diplom’s thesis introduced a novel generative model architecture for high-quality real space reconstruction of X-ray holography. His main efforts involved pattern recognition, uncertainty analysis, and manufacturing. Since June 2024, he has been a PhD candidate at the Faculty of Civil Engineering and Geosciences, TU Delft, focusing on uncertainty quantification for critical components using probabilistic deep learning approaches. His research aims to bridge conventional probabilistic methods and advanced deep learning and translate these tools to manufacturing to improve quality, reliability, and interpretability; his work is funded by the APRIORI doctoral network and TU Delft.

Renan Barros

Renan was born and raised in Fortaleza, Brazil. He obtained his bachelor's degree in Civil Engineering in 2021 and his master's degree in 2023, both from the Universidade Federal do Ceará. His master's thesis proposed a novel NURBS-based general Quasi-3D kinematic theory to account for both material and geometric nonlinear effects in functionally graded plates. His past research efforts also include Isogeometric Analysis, other kinematic theories and multiple types of analysis. Renan started his PhD in TU Delft in May 2024 in the position "Uncertainty quantification of physics-based and machine learning models of critical industrial components", as Doctoral Candidate 2 of the Horizon Europe DN APRIORI project, linked to the Marie Sklodowska Curie Actions program.

Siqi song

Siqi Song comes from China. After obtaining her master degree in offshore and dredging engineering at TU Delft, she started her career in Goldwind as a load and hydrodynamic engineer in 2021. During her work in Goldwind, she generated the integration analysis programming for offshore wind turbine substructure design and received the Value Creation Award in Goldwind in 2022 and developed multibody dynamic load models and tools for offshore wind turbine projects in China and Vietnam. She also participated in the DNV’s JIP ”ACE 2” project on behalf of Goldwind. In 2024, She shifted her focus to nacelle design and contributed to the development of the 16MW class offshore wind turbine design. After almost 4 years of industry experience, she decided to pursue an academic path and joined TU Delft as a PhD candidate in May 2025. Her research is experimental characterization and modelling of friction-related uncertainties for offshore applications. She is the third Doctoral Candidate in the APRIORI international consortium, which is funded by the Marie Skłodowska-Curie Actions program.

Maedeh Kabiripoor

Maedeh Kabiripoor is a mechanical engineer from Iran with a strong background in acoustics and metamaterials. After completing her bachelor’s and master’s studies at Isfahan University of Technology, where she explored how geometry and material properties influence sound properties, she gained practical experience in sound and audio testing in the home appliance industry. Currently, she is focusing on the industrial application of 3D-printed acoustic metamaterials as a doctoral researcher at KU Leuven within the Horizon Europe APRIORI project. Her research focuses on how variations in manufacturing geometry, density, and stiffness impact low-frequency sound absorption. Beyond models and lab tests, she is also exploring how these structures can actually be manufactured, with a close examination of real manufacturing methods, such as injection molding, to work towards designs that not only perform well in theory but can also be produced reliably on a larger scale.

Lorenzo Seghesio

Lorenzo, born and raised in Aosta, Italy, holds a Master's degree in Mechatronic Engineering (2022) and a Minor in Mechanical Engineering (2020) from Politecnico di Torino. During his Erasmus experience at California State University, Los Angeles, he completed his Master's thesis titled "System Integration, Localization, and Control Implementation for an Autonomous Mobile Service Robot." After graduation, he worked for a year and a half as a software design engineer for automotive applications at a company in Turin.

In January 2025, Lorenzo began his PhD at KU Leuven with the title "Quality Control within Injection Moulding Exploiting Data and Physical Models". He is the Doctoral Candidate number 5 in the APRIORI international consortium, funded by the Marie Skłodowska-Curie Actions programme.

Emily E Carvajal C

Emily E Carvajal C is a Research Engineer at Materialise HQ and enrolled as a PhD student at KU Leuven as DC6 within the framework of the Marie Curie Fellowship for the APRIORI project. Her main research interest is the use of AI/ML techniques for the analysis of multimodal and multidimensional data in multidisciplinary contexts. Her PhD research is focused on the use of data-driven models for first-time right metal additive manufacturing in part production workflow. The research is supervised by Dr. Michele Pavan and Professor Frank Naets.

Benham Firoozi

Behnam is a Marie Skłodowska-Curie PhD Fellow in the Solid and Computational Mechanics group at Aalborg University, Denmark (since December 2023), where he studies vibration and wave propagation in thin-walled structures. Previously, he was a research assistant at Kharazmi University in Tehran (2014–2019), focusing on nonlinear analysis of MEMS devices and energy harvesters. In 2019, with fellowship support, he joined the Vibration and Acoustic Laboratory (VAL) at Özyeğin University in Istanbul (2019–2023) and earned an M.S. in 2021. His master’s thesis examined vibration-based damage detection in mechanical structures, and at VAL he contributed to projects in piezoelectric energy harvesting, optimal microchannel heat-sink design, deep-learning energy forecasting, and unbalance detection in rotating machinery.

Hirenkumar Patel

Hiren Patel received his Master’s degree in Applied Mechanics from IIT Madras in 2020, specializing in Reduced-Order Modeling and vibration control of deterministic and stochastic systems. He gained experience in FEA and active vibration control at ARDB–DRDO, and later worked at Caterpillar applying machine learning for fault diagnosis and ROM for drivetrain components. Since Feb. 2024, he has been pursuing a PhD at Aalborg University with Grundfos, focusing on digital twins and reduced-order models for real-time failure prediction of mechanical components, funded by APRIORI.

Iman Hosseini

He holds a master's degree from Ural Federal University in Russia, where he applied machine learning to predict labor migration and optimize educational systems. Following his studies, he worked as a developer and programming instructor in Saint Petersburg, Russia, specializing in Python, C#, PostgreSQL, and MS SQL Server. In this role, he trained professionals across companies and educational institutions, strengthening their programming expertise. Earlier, after earning a bachelor's degree in computer engineering, he built a strong foundation as a Software Developer in Iran. Over the course of more than 15 years, he contributed to a diverse range of projects, including payroll systems for government agencies, educational software for telecommunications, hotel management platforms, and purchase and sales applications, consistently enhancing automation, efficiency, and user experience. His work also involved integrating payment systems, inventory management, and reporting tools to streamline business operations.

Currently, he is pursuing a Ph.D. at the Jožef Stefan Institute as part of the Marie Curie Fellowship within the APRIORI project. His research centers on active learning methodologies and strategies to enhance the explainability of machine learning models. By advancing transparency and interpretability, his work seeks to provide deeper insights into classification outcomes while reducing reliance on manual data labeling, contributing to more efficient and trustworthy AI systems.