
What happens to AI systems when they leave the lab and enter real production environments? This was the central question at the "AI in Production" workshop at KI2025 in Potsdam. The proceedings have now been published.
Michael Banf and Johannes Kuhn represented Perelyn, contributing a paper on the Model Context Protocol (MCP) in manufacturing. MCP is a standardised protocol that connects AI applications with external data sources and tools, comparable to USB-C for device connections. In the paper, our team examines what this means in practice for predictive maintenance, production planning and process automation.
Research at heart. Business in mind.
News

Dominik Filipiak and Michael Banf are co-authors of a community paper on the 2025 Topological Deep Learning Challenge, published in the Proceedings of Machine Learning Research. Their contributions feed into TopoBench, an open benchmarking library for the research community.
Event

Christian Mader attended Light + Building in Frankfurt for Perelyn – the world's leading trade fair for lighting and building services technology. The focus: how AI can advance intelligent building systems and connected infrastructure.