Pharmacist develops AI device to curb fish mortality

Artificial intelligence. Photo: Jobaaj Learnings

A Nigerian pharmacist, Ukachi Benita, has developed an artificial intelligence-powered device designed to help fish farmers detect dangerous changes in pond water conditions and reduce fish mortality.

 Benita developed the prototype, known as Aquamanne, during a Google Gemma hackathon after repeated fish losses on her mother’s farm raised concerns about how poor water quality affects fish production.

 According to TechCabal, more than 3,000 fish died in one of her mother’s ponds, which contained about 10,000 fish, in March 2026. The incident prompted Benita to investigate the possible causes of the losses and explore ways to monitor pond conditions more effectively.

 Aquamanne uses sensors and an ESP32 microcontroller to measure water parameters, including dissolved oxygen, pH, and temperature. The readings are processed by Google’s Gemma artificial intelligence model, which Benita ran locally during the hackathon, before being relayed to an application on a farmer’s phone.

 The system is designed to interpret water-quality readings and alert farmers when conditions change, while providing recommendations on steps they can take to reduce risks to their fish.

 Explaining the role of artificial intelligence in the innovation, Benita told TechCabal that the technology could help bridge language barriers by presenting information in simpler terms and eventually supporting text and voice communication in African languages.

 “The main thing AI does is address the language barrier,” she said. “A normal sensor might just give you the readings, but the conversion to local languages is not there; the voice capability as well is not there.”

 The initial prototype, which cost about N30,000 to build, was completed in a week using components sourced in Nigeria. However, some of the sensors Benita wanted were unavailable locally, forcing her to work with the parts she could obtain.

 The pharmacist, who studied at Nnamdi Azikiwe University, Awka, Anambra State, said she had limited experience with embedded systems before participating in the hackathon and relied on online tutorials and artificial intelligence tools while developing the device.

 The project was inspired by a wider challenge confronting fish farmers, who can suffer significant financial losses when fish die because of deteriorating water conditions.

 A 2018 study of 72 small-scale fish farmers in two local government areas found that 76.4 per cent identified water pollution as a major cause of disease on their farms.

• Trump asks followers to vote on new name for AI
• Why food prices remain high despite increased production – Minister
• Firm unveils plan to boost oil palm production in Akwa Ibom

 A catfish farmer in Bariga, Lagos, Sesan, who manages 10 ponds containing about 8,000 fish, also described how oxygen depletion caused by water pollution once led to the loss of about 30 per cent of his stock.

 “The fish started dying immediately after the oxygen depletion,” he said.

 Another fish farmer, Goodluck Juwan, recalled losing 5,000 catfish in 2021 after beginning his farming journey with 500 fish in 2017. He later acquired a device for measuring temperature and pH and began monitoring his ponds more regularly.

 Juwan said a system that could monitor water conditions and advise farmers on what action to take would be valuable to fish producers.

 “If there is a tool like that, I want to have it,” he said. “A lot of fish farmers would pay to have it.”

 Following the hackathon, Benita told TechCabal that fish farmers who saw her posts on social media began contacting her about their own challenges. One farmer with a fish farm in Owerri, while living in Lagos, asked whether the system could enable him to monitor his pond remotely.

 The feedback encouraged Benita to move beyond the prototype and work towards developing Aquamanne into a product for wider use.

 She is now working with two collaborators, a Nigerian professor based at Oxford and an embedded systems engineer, to improve the device.

 The team plans to make the hardware smaller and more durable, reduce its dependence on electricity, improve remote monitoring capabilities and enhance its support for African languages.

 Benita said the team also wanted to keep the device affordable and easy for farmers to install and operate.

 “We want to use the best hardware systems, but we don’t want it to be at the cost of the farmers,” she said.