The technology was developed two years ago. Nkundimana told IGIHE that he began working on it while at university, where he studied agricultural technology, with the aim of developing solutions to challenges facing the sector.
Agric AI relies on crop-related information available on its website and mobile application.
To use the technology, a farmer opens the application on a smartphone and points the phone’s camera at a crop leaf. The AI then compares the image with information already in its database, analyses the findings and provides feedback to the farmer.
“The technology immediately shows the condition of the crop, whether it needs treatment or has another problem, and helps the farmer understand what action to take,” Nkundimana said.
However, because some of the information provided is technical, farmers can also consult agricultural extension officers or other agricultural experts for further guidance on the appropriate action to take.
Currently, the technology provides results with an estimated 70 percent accuracy. More data is still being added to improve the accuracy of its analysis.
The platform currently contains information on 11 crops, although it is being used to assess four crops so far: coffee, tomatoes, beans and maize.
About 250 farmers are currently using the technology, which remains free of charge. Nkundimana said the company may introduce a payment model as the number of users grows and the technology becomes more fully developed.
Explaining how the innovation could benefit the agricultural sector, he said access to reliable information has been one of the biggest challenges facing farmers.
“This is a solution because access to information has been one of the main challenges for our farmers. A farmer might have a crop with a problem but not know what it is. They would go to an agricultural pharmacy to buy treatment without being certain what problem they were trying to solve. We want farmers to go there knowing exactly which treatment matches the problem affecting their crops,” he said.
The technology could also benefit investors and other players across the agricultural value chain. By collecting and analysing information as it becomes available, it could help identify areas facing challenges and support better-informed investment decisions.
Nkundimana said he currently works with a team of 10 people who help develop the technology, although not all of them are permanent, salaried employees. He added that the company still needs funding to expand its operations.
He said he hopes to acquire a server capable of storing and processing much larger volumes of crop data. In the future, the platform is expected to go beyond crop diagnosis and also assess soil conditions, helping farmers adopt more modern and data-driven farming practices.
Nkundimana added that the company still needs information and feedback from farmers about their crops and farming experiences, noting that AI cannot perform effectively without sufficient data. He therefore urged farmers to continue sharing information about their crops to help improve the technology.
