Sadako Technologies won the $100,000 prize at Nvidia’s Emerging Companies Summit with a robot that uses machine learning to sort recycled products.

The Barcelona, Spain-based Sadako beat out 11 other contestants in the annual Nvidia Early Stage Challenge at the Emerging Companies Summit at Nvidia’s GPUTech conference in San Jose, California, yesterday. The event was a Shark Tank-style presentation with a panel that included tech pundit Rob Enderle; George Hoyem, investment partner at In-Q-Tel; Brandon Farwell, a partner at Rothenberg Ventures; and Jeff Herbst, vice president of business development at Nvidia.

Sadako has created a robot that uses machine learning to sort recycling and then extract specific types of valuable waste, like PET plastic bottles.

Eugenio Garnica, Sadako’s CEO and cofounder, used his allotted four-minute presentation slot during the event to¬†describe how his robot differentiates types of trash and then snatches it with a specially designed gripper. While only three units have thus far been sold, he said the robot has a payback period of under 16 months, and it’s seeing¬†strong interest in a market worth up to $3 billion.

Nvidia received more than 100 applications for the contest from a half-dozen countries. The contest supports companies that use Nvidia’s graphics processing units (GPUs) in clever ways.

The other Early Stage Challenge competitors were:

  • Aerialguard (Israel) — Provides autonomous situational awareness for drones and UAVs, dramatically increasing safety, survivability, and mission capabilities.
  • CogniCor (Spain) — Uses AI and natural language processing for handling product queries, claims, and other customer support issues.
  • Lucid VR (U.S.) — Develops LucidCam, a stereoscopic 3D camera for consumers with 180-degree wide-angle lenses and spatial audio.
  • Linkface (China) — Offers facial recognition technology powered by deep learning. Its cloud platform provides free services for noncommercial use and high-concurrency services for business use.
  • Intelligent Voice (U.K.) — Offers speech-to-text technology and analysis of unstructured communications for compliance purposes, including the collection, processing, and analysis of audio and other data types.
  • Horus Technology (Italy) — Develops a wearable device that uses computer vision and machine learning to aid visually impaired people, describing the environment through bone conduction.
  • Hypercubes (U.S.) — Develops satellites that reveal unprecedented details of Earth, with the ability to remotely classify chemical compositions for applications such as precision farming, mining, and oil and gas operations.
  • BriSky Technology (China) — Develops all-weather industrial drones that use computer vision and deep learning to conduct tasks such as autonomous inspections of power lines and wind turbines, traffic monitoring, surveillance, and public security.
  • TempoQuest (U.S.) — Develops software as a service to meet the need for faster, more accurate weather forecasts for commercial users and government agencies.
  • Entropix (U.S.) — Enables inexpensive cameras like those in smartphones and drones to capture extremely high resolution images at up to 8K.
  • Analytical Flavor Systems (U.S.) — Uses machine learning and AI to identify and predict in real time flaws, contamination, and batch-to-batch deviations for food and beverage producers.