Berlin Metropolitan Area
CTO, physicist, and researcher focused on turning data and AI into reliable products and platforms in complex, regulated environments. As Partner & CTO at Ambrosys, I build platforms, AI capabilities, and engineering organizations in tolling, mobility, energy, and regulated environments. My background spans physics, machine learning, and software engineering, from academic research to large-scale production systems. My work sits at the intersection of technology, product, data, and organizational design: platform architecture, MLOps, streaming data systems, and applied AI that performs under real-world constraints.
Ambrosys is a technology and consulting company building AI-powered, data-driven solutions for the mobility, tolling, and energy sectors. As CTO, responsible for tech strategy, product direction, engineering org growth, and agile software delivery.
Led a small engineering team to develop a test case management system for map-matching algorithms using multiple geospatial data sources. Focused on quality assurance and reproducibility in mobility applications.
Led and contributed to applied machine learning projects in collaboration with universities and industry partners. Focus areas included time series prediction, map-matching algorithms, and spatial data systems. Bridged scientific research and engineering by developing prototypes, publishing findings, and supporting grant-funded initiatives.
A community network of CTOs, tech leads, and forward-thinkers who drive innovation and vision.
Co-founded a platform venture for MLOps and applied machine learning. Shaped the product vision, built early prototypes, and transitioned the work into ongoing research and development at Ambrosys.
Developed the MVP and core infrastructure for a machine learning–based wind energy forecasting platform. Designed and implemented microservice architecture, data pipelines, and a prediction engine, laying the technical foundation for a successful spin-off in the renewable energy sector.
DAAD-funded research stay with the Brunton/Kutz lab, focusing on sparse regression and interpretable modeling of dynamical systems. Initiated and co-authored the open-source library PySINDy, now widely used in the scientific and applied ML communities. Contributed to collaborative publications on data-driven system identification and control.