AgroguardAI

Inspiration

I grew up in a farming community in Nigeria where smallholder farmers regularly lose large parts of their harvest to pests they cannot identify early. Most existing solutions assume good internet access, high literacy, and expert knowledge, which many farmers do not have. AgroguardAI was built to provide practical AI support that works in real field conditions.

What it does

AgroguardAI is a farm decision system that uses Gemini 3 to diagnose crop pests and diseases from images and voice inputs. It then gives clear, actionable treatment and timing advice that farmers can easily understand and use.

How we built it

Gemini 3 is the core reasoning engine of the system. We use its multimodal capabilities to analyze real crop images and spoken farmer questions, even in noisy outdoor environments. Through carefully designed system prompts, Gemini is transformed into a context-aware digital agronomist that produces structured diagnoses and practical recommendations instead of generic text.

Challenges we ran into

Farm data is messy. Images are often low quality, and voice input includes strong accents and background noise. Aslo language barrier problem we're working hard to see everything is okay.

Accomplishments we’re proud of

We built a working MVP in two months, achieved about 90% accuracy on common pests, and demonstrated the system to farmers and government ICT officials.

What’s next for AgroguardAI

We'll seek investment/Grants to fund R&D, including extending Gemini-powered diagnostics to low-cost drone scouting, simple farm robota and expanding validation across more crops and regions.