DTU Compute at the Technical University of Denmark is hiring a fully funded PhD student to develop machine learning methods for X-ray nano-tomography, working with research groups at PSI and EPFL in Switzerland.
The role
- Develop machine learning and deep learning methods that jointly reconstruct the sample structure, the acquisition geometry and the sample deformation from inconsistent projection data.
- Study neural field representations for 3D and 4D imaging and combine learning-based methods with physical models of image formation.
- Validate the methods on experimental nano-tomography data, starting with integrated circuits.
- Collaborate with DTU, PSI and EPFL; research visits to Switzerland are planned.
- Publish in leading journals and conferences and help supervise BSc and MSc projects.
Requirements
- A two-year master’s degree (120 ECTS) or equivalent in computer science, applied mathematics, physics, engineering or a related field.
- A strong background in machine learning and deep learning, with experience developing, implementing or evaluating learning-based models.
- Scientific programming, preferably Python, and a solid understanding of mathematical modelling, optimisation or numerical methods.
- Advantages: neural representations, 3D vision, inverse problems, image registration, tomographic reconstruction.
- Good written and spoken English. You may apply before finishing the master’s degree but cannot start before receiving it.
- The ad says open-source background checks may be run on qualified candidates, because the work falls under security and export-control rules.
About the employer
DTU Compute is DTU’s department of applied mathematics and computer science, with over 400 employees, 10 research sections and about 140 PhD students. The project is linked to the Center for Quantification of Imaging Data from MAX IV (QIM).
Salary and deadline
- Three years, full time, starting 1 March 2027 or by agreement.
- Pay follows the collective agreement with the Danish Confederation of Professional Associations; the ad gives no amount.
- Deadline: 16 November 2026, 23:59 Danish time. Apply in English with one PDF containing a cover letter, CV, transcripts and BSc/MSc diplomas with the grading scale.
Source: DTU Compute, Technical University of Denmark, “PhD Scholarship in Machine Learning for Non-Stationary and Non-Static Nano-Tomography – DTU Compute”, read on 25 October 2026. Summary by AVIDE; see the original page for the full, current text.