University of Seoul student wins agriculture category of World Bank Group AI hackathon
Summary
Moon Tae-yoon, a University of Seoul student, won first place in the agriculture track of the World Bank Group’s Small AI for Development Hackathon Challenge. Held from Oct. 3-4, the competition sought practical AI solutions for places with limited internet access and electricity. Moon developed Dure, which helps smallholder farmers coordinate collective sales using basic mobile phones. He later demonstrated the solution and presented its technical architecture at the Global AI & Digital Summit 2026 in Seoul from Oct. 19-22.
Key Facts
- Moon competed individually, while teams from the University of California, Berkeley, and Carnegie Mellon University placed second and third, respectively.
- Dure coordinates collective shipments, facilitates buyer bidding and arranges joint transportation through text messages sent from basic mobile phones.
- The system combines an ultralightweight language model with a rule-based engine to process text messages in Tetum, one of Timor-Leste’s official languages.
- Dure can build a database of registered farms from transaction records and generate data and recommendations to support policymaking.
AI solution aims to help smallholder farmers in developing countries

Moon Tae-yoon, second from left, a student in the Department of International Relations at the University of Seoul, poses with Kim Sang-bu, third from left, the World Bank Group’s vice president for digital and artificial intelligence (AI), and other participants at the Global AI & Digital Summit 2026, held in Seoul, Oct. 19-22. Courtesy of University of Seoul
Moon Tae-yoon, a student in the Department of International Relations at the University of Seoul, has won first place in the agriculture track of the Small AI for Development Hackathon Challenge hosted by the World Bank Group (WBG).
Held from Oct. 3-4, the hackathon was designed to identify practical artificial intelligence (AI) solutions that can be deployed even in environments with limited internet access and electricity infrastructure.
Competing as an individual, Moon presented “Dure,” an AI solution he independently conceptualized and developed.
A team from the University of California, Berkeley, took second place, while a team from Carnegie Mellon University finished third.
“The knowledge of development cooperation I gained while studying international relations, along with the practical experience I acquired while participating in a KOICA [Korea International Cooperation Agency] agricultural value chain project in Timor-Leste, provided a strong foundation for developing AI-based solutions to real-world challenges,” Moon said.
“The experience was particularly meaningful because I was able to go beyond simply developing technology and propose practical solutions that could be applied in the field,” he added.
In recognition of his achievement, Moon was officially invited to the Global AI & Digital Summit 2026 (GADS 2026), co-hosted by the WBG, Korea's Ministry of Economy and Finance, and the Ministry of Science and ICT.
At the summit, held in Seoul from Oct. 19-22, he operated an exhibition booth where he demonstrated Dure to representatives of governments and international organizations from around the world.
He also participated as a speaker in an official session, presenting the project’s technical architecture and its potential to drive innovation in agricultural markets in developing countries.
Inspired by Korea’s traditional rural community organization known as “Dure,” a system of mutual assistance among farmers, the solution is designed to help smallholder farmers in developing countries strengthen their bargaining power through collective transactions without having to establish formal cooperatives.
Its key functions include coordinating the collective shipment of agricultural produce, facilitating buyer bidding and arranging joint transportation through text messages sent from basic mobile phones.
When selling their produce individually, smallholder farmers in developing countries often struggle to negotiate favorable prices with buyers.
Establishing formal cooperatives, however, can involve significant barriers, including administrative procedures, initial capital requirements and the need to reach consensus among members.
Dure aims to overcome these constraints by enabling smallholder farmers to participate in collective transactions, thereby delivering key benefits of cooperatives even in environments with limited digital infrastructure.
The Dure project was developed with agricultural communities in developing countries in mind, including coffee farmers in Timor-Leste.
It combines an ultralightweight language model with a rule-based engine to process text messages in Tetum, one of the country’s official languages. The solution is designed to operate without requiring users to install a separate application or maintain a constant internet connection.
It can also build a database of registered farms using transaction records and generate data and recommendations to support policymaking.
Explore More
- Q.
- Q.
- Q.