A startup led by an LSU alumni and two current students is making strides in transforming how farmers and agricultural professionals access essential information.

FarmSmart, an AI-powered app currently in beta testing, promises to streamline complex research from dense agricultural documents into quick, user-friendly responses available at the touch of a button.

The idea for FarmSmart came about after then-LSU student Colin Raby had a conversation with Professor David Stevenson at the LSU AgCenter. Stevenson, who often fields questions from young farmers, noticed a common trend: many of these farmers weren’t consulting the AgCenter’s annual guide – a comprehensive, 500-page document detailing best practices in crop management and pest control.

Stevenson would encourage the farmers to check the guide, but many found the manual challenging to navigate. Although the guide is a valuable resource, its dense, black-and-white text made it less accessible to those seeking quick answers.  Raby realized there was an opportunity to make this information easier to access.

“If you had a huge user manual, what would you prefer: to read through the entire thing or just get your answer?” Raby asked.

In 2023, Raby and FarmSmart co-founder Grant Muslow took a course at LSU on large language model technology, which uses artificial intelligence to understand and respond to user inquiries. For a class project, they decided to build an app specifically for LSU AgCenter documents, which soon grew into the ambitious, AI-enabled FarmSmart.

The app now aims to help not just Louisiana farmers but also agricultural professionals nationwide. In the beta testing phase, it is programmed to identify and solve crop issues across 15 states and is planned to expand to 50 states in the future, according to Raby. 

FarmSmart works by analyzing thousands of agricultural documents, including research from the EPA and the Department of Agriculture, as well as LSU-specific resources. Its Louisiana model is currently the most accurate, but the team plans to expand accuracy to additional states once they finalize the Louisiana version.

FarmSmart’s AI is also transparent about its limitations, according to Raby. If it encounters a question it can’t answer, the app simply says it doesn’t know, rather than offering potentially inaccurate information. Additionally, the app provides citations for its responses to show users where information is sourced.

“Our goal is to match or exceed the quality of the best agricultural consultants,” Raby said.

While hiring agricultural consultants can cost farmers hundreds of thousands of dollars depending on their farm’s size, FarmSmart aims to provide comparable information at a fraction of that cost, although the exact pricing has yet to be determined.

The FarmSmart team consists of Raby, Muslow and Julius Pallotta, a biological engineering major at LSU who handles the front-end code. Raby serves as CEO and CFO, overseeing both the technical aspects and the business side, while Muslow, a graduate student in electrical engineering with a concentration in machine learning, is the chief technology officer. Pallotta, currently on a semester break, serves as the chief operations officer.

Raby also currently serves as a Congressional Artificial Intelligence Specialist. 

The team has already onboarded 15 beta testers, including local farmers, agronomists and landscape architects. Their feedback is helping the team identify strengths and potential improvements, particularly in making the user interface more intuitive. FarmSmart’s goal is to reach 50 beta testers and continue refining the app’s performance.

Farmers interested in agriculture who want to try the Agricultural Assistant for free are encouraged to become beta testers and provide valuable feedback on its functionality. To apply, visit the FarmSmart website at www.FarmSmart.ai and click on the “Become a Beta Tester” button. Alternatively, you can access the application form directly through this link:Beta Tester Application.