Hi there :)
I'm a physicist working on my phd at Laser Zentrum Hannover.
I usually end up building both halves of an experiment: the instrument that makes the data, and the software that turns it into an answer. That means readers and converters for raw instrument output, the pipelines that feed a trained model or trains it, the evaluation that decides whether the result is worth anything, and an interface the whole thing can be operated from by someone who wasn't there when it was written.
Python, PyTorch, PyTorch Lightning, Optuna, Weights & Biases, NumPy/SciPy, CUDA. Fluorescence-lifetime imaging, time-correlated single-photon counting, phasor methods. Semantic segmentation, microscopy image analysis, scientific data pipelines. FastAPI, Docker, Linux, and plenty of Git.
P. Dyrøy, J. Heitz, H. Studier, S. Johannsmeier, T. Ripken. Phasor-based FLIM analysis of NAD(P)H and FAD autofluorescence for label-free bacterial classification. Journal of Biomedical Optics 31(1), 016502 (2026). doi:10.1117/1.JBO.31.1.016502
Most of what I've written lives in private repositories: phd work that isn't published yet, and code that belongs to the institute rather than to me. Pieces get split out and released as the corresponding papers appear. Happy to walk through any of it in a conversation.