Russian Students Build AI Apps to Predict Crop Yields and Test Seed Quality
At a national hackathon, young developers used machine learning and data analytics to tackle real problems in agribusiness.
Students from 28 Russian universities have created digital prototypes designed to transform the country’s agricultural sector. The projects emerged from the Technoforge: AgroTECH hackathon, held at the Moscow Institute of Electronics and Mathematics (part of the Higher School of Economics). The event’s co-organizer, agribusiness giant EkoNiva, plans to help test and deploy the winning projects across its farms.
From digital models to real-world fields
Among the standout ideas was EKO CONTROL, a data-driven service for soil analysis and crop yield forecasting. Using detailed fertilizer maps and a simulation engine, it evaluates how weather and agricultural chemicals affect harvest outcomes.
Another team, called “Agroschoolers,” built an AI-powered mobile app that checks seed quality using Vision Transformer technology — the same neural network architecture behind many leading image-recognition systems.
A third promising prototype uses computer vision and machine learning to identify plant growth stages from field images, helping farmers fine-tune irrigation and fertilization schedules.
Winning teams are now preparing pilot programs to test their solutions in real agricultural conditions — a step toward digitizing Russia’s vast farmland and making its harvests more predictable, efficient, and sustainable.
