I study computer science at Leipzig University and work on machine learning. I started out mostly on the engineering side, building systems and pipelines. Over time, I became more interested in the research questions behind them: why a method works, what information it uses, and where it fails.
I’m especially interested in graphs, time series, and machine learning: how to represent structure and dynamics, and what models can learn from them.
This is an early version of my site. I plan to add project pages, notes, and a few interactive examples. Some sections are still missing.
Right now, I am working on my master’s thesis and, in parallel, as a research assistant at InfAI.
My thesis uses the rigid equal-sphere clusters enumerated by Miranda Holmes-Cerfon. Starting from a contact graph, I study how machine learning can help construct a corresponding three-dimensional realization and what geometric information a model learns in the process.
Since May 2026, I have been working on the thesis as a visiting researcher at the Max Planck Institute for Mathematics in the Sciences in Leipzig.
Alongside my thesis, I work as a research assistant at InfAI on the KORALLE project.
A few earlier stops that shaped the questions I work on now.
Research internship on real-time anomaly detection for multivariate time series, with a focus on efficient inference and reproducible evaluation.
Semester abroad in computer science; named to the Dean’s List.
Research assistant in the HyGraph project, working on data models that connect temporal graphs and time series.
Most of my work involves structured data. I want to understand how relationships and temporal patterns can be represented, and what models can learn from them.
I want to use it to document my work properly, with enough context to understand the problem, the method, and the parts that did not work.