About Karevo
Optical sorting technology for potatoes and onions – Affordable precision for family farms and direct marketers.
Karevo develops a sorting machine for optical quality control of table potatoes, seed potatoes, and onions. Powered by artificial intelligence, the Duo85 detects seven different defects: greening, growth cracks, dry rot, wet rot, wireworm, scab, and mechanical damage, as well as foreign bodies such as stones and clods. With a throughput of up to 5 tonnes per hour and a detection accuracy of approximately 95%, the system relieves farms of the burden of manual sorting.
The Problem
On small and medium-sized farms, quality sorting is often still done by hand. This process requires a great deal of labour and leads to high personnel costs. The work at the sorting table is monotonous, physically demanding, and characterised by a loud and dusty environment. Existing sorting systems on the market are frequently designed for large industrial operations and cost significantly more than Karevo's machine.
Our Solution
Karevo's innovation lies in the use of artificial intelligence for image processing and the compactness of the system. The camera system, the pneumatic ejection mechanism, and the control unit can each be replaced individually. Image recognition runs locally on a powerful edge device, independent of an internet connection. Via an intuitive display, the farmer can individually adjust the AI settings and define which defects are sorted out.
Karevo is a spin-off of the Technical University of Munich (TUM), founded in 2024 by Benedikt Keßler, Johannes von Wittke, and Felix Beck. The venture is based on Benedikt's Master's thesis in mechanical engineering at TUM on the automated quality control of unwashed table potatoes using computer vision and machine learning. Benedikt grew up on a potato farm and experienced the lack of affordable automation technology first-hand. The Duo85 is the machine he wished he had for his farm.
Awards and Memberships
- Best Agri-Tech Startup, Growth Alliance (2025)
- Future Agro Challenge Germany (2024)
- 3rd Place AI+ Munich (2024)
- Member of the German Agribusiness Alliance (since 2026)
- Funded by the EXIST Start-up Grant of the BMWK