Research

My research bridges machine learning, physics, and industrial applications, with a focus on creating interpretable, reliable AI systems for semiconductor manufacturing.

Research Pillars

Scientific Machine Learning for Metrology & Inverse Problems

Developing physics-informed and hybrid machine learning models that combine domain knowledge from simulations with experimental signals. This research focuses on solving inverse problems in semiconductor metrology, uncertainty quantification, and creating reliable surrogate models for complex physical systems.

Physics-Informed Neural Networks Hybrid Modeling Uncertainty Quantification Inverse Problems Surrogate Models

Interpretable & Trustworthy ML

Building machine learning systems that are transparent, interpretable, and reliable for high-stakes applications in semiconductor manufacturing. This work emphasizes uncertainty quantification, model interpretability, and auditing capabilities to ensure AI systems can be trusted in critical decision-making processes.

Model Interpretability Uncertainty Quantification Trustworthy AI Auditable Systems Reliability Engineering

Data-Driven Discovery in Physical Sciences

Leveraging machine learning and data mining techniques to uncover composition-property relationships and discover new patterns in physical sciences. Drawing on a computational physics background, this research applies scientific ML to materials science, defect inspection, and metrology challenges.

Materials Informatics Defect Inspection Scientific Computing Data Mining Computational Physics

Collaboration

I actively co-supervise MSc and PhD students in collaboration with Dutch universities, working on projects at the intersection of machine learning, physics, and semiconductor technology.

Interested in collaborating on research projects, student supervision, or industrial partnerships? I'm always open to discussing new opportunities in scientific machine learning and its applications.

Get in Touch