Moritz Stammel
AI researcher and engineer
Updated: October 2026.
I’m interested in how AI systems can learn clinically meaningful representations, remain reliable across changes in data and practice, and reveal their blind spots before those failures affect care.
Research
My research sits at the intersection of representation learning, generative modelling, and robustness. I’m interested in how models of imaging and biology can help us understand how systems generalise across hospitals, scanners, and patient populations.
My recent work investigates test time counterfactuals for stress testing medical imaging models under controlled distribution shifts and comparing their behaviour with real data from outside the training distribution.
More broadly, I’m interested in representation learning, including how models can learn shared representations across modalities, and in generative modelling as a way to capture structure in complex data.
I completed an MSc in Advanced Computing at Imperial College London, after studying Computer Science at TU Dresden.
Engineering
At QuantCo and VIRDX, I work on machine learning research and development for medical imaging. At VIRDX, I work on getting histopathology level insights from MRI images using biophysical simulation and inverse modelling.
Public engagement
I’m fascinated by how technology shapes society and by who benefits from the systems we build. I’m active in the Studierendenforum im Tönissteiner Kreis, where I serve on the executive board, and cofounded the Rethinking Healthcare initiative. I love to bring people together in a room to discuss various topics and exchange ideas.
Elsewhere
<!-- thanks for reading. -->