Karevo, a startup founded in 2024, has developed an AI-powered optical recognition system that sorts potatoes at up to 10 tons per hour with 95% accuracy. The machine can identify damage and defects including rot, cracks and wireworms, and also detects foreign objects. According to the company, the system works effectively on unwashed potatoes, a capability that other detection methods often lack.
Co-founder Benedikt Keßler grew up on a potato farm and developed the concept while studying mechanical engineering at the Technical University of Munich (TUM). He worked with co-founders Johannes von Wittke, an engineering student at the East Bavarian University of Applied Sciences in Regensburg, and Felix Beck, who completed his Executive MBA at TUM. The AI model was trained on more than 100,000 images.
Keßler said that potato sorting machines are often too expensive, too large or too complicated to maintain for small farms. Karevo designed its machines to be smaller and more affordable, with a modular design that simplifies maintenance and allows integration into existing systems. The team participated in the UnternehmerTUM Incubator, a center for innovation and entrepreneurship at TUM, during the startup phase and brought the machines to market in fall 2025.
