Multiple papers accepted at top conferences such as NeurIPS 2025, ICML 2025, ICLR 2025; notable works include 'Computational Algebra with Attention: Transformer Oracles for Border Basis Algorithms', 'Neural Discovery in Mathematics: Do Machines Dream of Colored Planes?', and research on model compression, federated learning, etc.
Research Experience
Currently serves as DL Research Lead at the IOL Lab of Zuse Institute Berlin; before joining IOL in 2020, worked on combinatorial optimization problems with Leon Sering at the COGA Group at TU Berlin.
Education
A 4th-year PhD candidate in Mathematics at TU Berlin, advised by Prof. Dr. Sebastian Pokutta. Since 2022, a member of the BMS graduate school, part of the MATH+ Cluster of Excellence. During BSc and MSc studies in Mathematics at TU Berlin, interned in research groups led by Prof. Sergio de Rosa at Università degli Studi di Napoli Federico II and Prof. Marco Mondelli at IST Austria.
Background
Research interests include enhancing the compute and memory efficiency of large Neural Networks through sparsity and quantization; also interested in applying Deep Learning to scientific discovery (AI4Science/AI4Math), addressing sustainability challenges, or within the context of Agentic AI. Other areas of interest include Federated Learning, XAI, and Optimization.
Miscellany
Shares TLDRs of some papers on his blog; team is actively recruiting motivated PhD students for work on Deep Learning.