We are seven people in Munich. Between us: a decade of research on multi-agent reinforcement learning and planning at LMU Munich and TH Ingolstadt, and years of building data and AI systems in industry, at companies like MaibornWolff, FINN, BMW and Siemens. We have shipped agents to production before, and we have seen where they fail.

That is why our agents run on your infrastructure, act within the limits you set, and are judged on one thing: how much less your operation wastes.