Summary
The Advanced Research Center for Nanolithography (ARCNL) is offering a fully funded 4-year PhD position focusing on combining physics-informed machine learning and computational modeling to reconstruct nanoscale structural information from imperfect, low-resolution, and noisy metrology data. This research is conducted in close collaboration with ASML, the Centrum Wiskunde & Informatica (CWI), and the AI4Science Lab at the University of Amsterdam (UvA).
PhD Position: Physics-Informed Machine Learning for Semiconductor Metrology
Designation
PhD Researcher / Doctoral Candidate (Employed by NWO-I)
Overview & Quick Facts
| Parameter | Details |
| Host Institute | Advanced Research Center for Nanolithography (ARCNL) |
| Collaborators | ASML, CWI (Prof. Dr. Tristan van Leeuwen), UvA AI4Science Lab (Dr. Patrick Forré) |
| Employment Type | Full-time (40 hours/week, 12 months/year) |
| Duration | 4 Years |
| Gross Salary | Starting at ~€3,115/month (NWO-I collective labor agreement scale) |
| Benefits | Visa assistance, housing assistance for international candidates, relocation/furnishing expense compensation, 8% holiday pay, and end-of-year bonus |
Research Area
- Physics-Informed Machine Learning (PIML)
- Semiconductor Metrology & Nanolithography
- Inverse Problems & Computational Imaging
- Mathematical Modeling & Scientific Computing
- Optimal Experimental Design
Location
ARCNL (Amsterdam Science Park), Amsterdam, Netherlands
Eligibility & Qualifications
- Academic Background: An MSc degree (completed or near completion) in Computer Science, Artificial Intelligence, Machine Learning, Applied Mathematics, Physics, Computational Science, or a related discipline meeting Dutch university entry criteria.
- Core Competencies: Prior experience or strong background in machine learning, inverse problems, scientific computing, or data-driven physical modeling.
- Skills: Strong analytical and algorithmic problem-solving skills, collaborative mindset for multi-institute/industry projects, and fluency in written and spoken English.
Scholarship & Project Description
Modern semiconductor fabrication requires measuring nanometer-scale wafer structures with extreme precision and high throughput. This creates an inverse problem: inferring hidden physical dimensions from noisy, limited, and low-resolution data.
As a PhD researcher in this project, you will:
- Develop novel physics-informed machine learning frameworks that merge physical simulations of light-matter measurement processes with inverse reconstruction algorithms.
- Perform data-driven design of experimental setups, optimizing measurement strategies and parameters to maximize information extraction.
- Work across academic and industrial boundaries with leading experts at ARCNL, ASML, CWI, and UvA.
- Complete coursework, publish findings in peer-reviewed scientific journals, and defend a doctoral thesis for a PhD degree from a Dutch university.
How to Apply
- Visit the official job portal posting on ARCNL Jobs or via AcademicTransfer.
- Submit your application dossier through the online application button, including:
- A detailed Curriculum Vitae (CV).
- A Motivation Letter explaining your interest in physics-informed ML and relevant background.
- Transcripts of your BSc and MSc records.
- Contact details of 2 academic referees.
Last Date to Apply
- Open until filled / Rolling review: Applications are reviewed on an ongoing basis until a suitable candidate is selected. Prompt application is encouraged.
