FarmGuard AI

Inspiration

The inspiration for FarmGuard AI came from the realization that while AI is advancing rapidly, its benefits often bypass small-scale farmers in underserved communities. These farmers are the backbone of global food security, yet they face devastating crop losses—up to 40%—due to pests and diseases. We wanted to bridge this gap by providing an "agronomist in your pocket" that doesn't just diagnose problems but offers affordable, organic solutions tailored to marginalized farmers who cannot afford expensive chemical pesticides.

What it does

FarmGuard AI is a mobile-first platform that empowers farmers to: Identify Diseases: Use a smartphone camera to instantly detect crop issues like aphids or blight. Access Organic Remedies: Provides DIY, step-by-step recipes (e.g., Neem oil solutions) using locally available ingredients. Track Impact: A real-time dashboard visualizes the "Yield Saved" and "Environmental Savings." Calculate Economic Benefits: A built-in calculator helps farmers understand the financial value of their recovered crops.

Who it helps

FarmGuard AI is specifically designed for smallholder farmers in underserved or marginalized regions who operate outside the reach of traditional agricultural support systems. The Resource-Constrained: Farmers who cannot afford expensive synthetic pesticides or professional agronomist consultations. The Remote: Individuals in rural areas with limited access to agricultural experts but who possess a basic smartphone. Marginalized Communities: Low-income families whose entire livelihood depends on the success of a single acre of land, where a 20% crop loss can mean the difference between food security and poverty.

How we built it

We focused on creating a robust, full-stack experience: Frontend: Developed with React and TypeScript for a professional, type-safe user experience. Backend & Auth: We integrated Supabase to manage secure user authentication and store persistent farmer profiles. AI Engine: We utilized the Google Gemini 1.5 Flash model to analyze diagnostic data and generate personalized treatment recommendations.

Why it matters

Agricultural health is a cornerstone of global stability, yet the "digital divide" often leaves those who need technology most with the least access to it. Economic Sovereignty: By providing organic, DIY recipes, we move farmers away from a cycle of debt associated with purchasing expensive chemical inputs. Environmental Justice: Synthetic runoff from small farms often affects local water supplies in underserved areas; promoting organic alternatives protects the community's health. Democratizing Knowledge: We are taking state-of-the-art AI (Gemini 1.5 Flash) and making it a practical, daily tool for someone who may have never used an AI before.

How it creates measurable impact

We don't just guess that the app works; we use data to prove it. Our platform tracks specific metrics that translate directly into human impact: Yield Recovery Efficiency: We measure the delta in predicted crop yield before and after the application of an AI-suggested remedy. Chemical Reduction Metric: We track every litre of organic remedy used as a direct substitution for synthetic alternatives. Economic Resilience : We calculate the actual Dollar value saved by the Farmer Time-to-Diagnosis: By reducing the time to identify a disease from days (waiting for a scout) to 0.3 hours (instant scan), farmers can act before the infestation spreads, significantly increasing the probability of crop recovery.

Challenges we ran into

Environment Configuration: Securely managing API keys was a hurdle. We initially struggled with accidental exposure but eventually implemented a robust system using a gitignored config.ts to keep credentials safe. TypeScript Strictness: Moving to a strictly typed environment caused friction with initial JavaScript files (like the implicitly has an any type errors), but fixing these made the app significantly more stable. Data Privacy: Ensuring that marginalized farmers' data was private required learning how to implement Row Level Security (RLS) in Supabase so users only see their own farm logs.

Accomplishments that we're proud of

Measurable Impact: We successfully built a dashboard that quantifies real-world change, such as 0.8L of chemical runoff prevented. Clean UI/UX: Designed an interface that feels accessible and modern, specifically for users who may not be tech-savvy. Full-Stack Integration: Successfully syncing Supabase Auth with a custom PostgreSQL "Profiles" table to store farm details.

What we learned

We learned that innovation for impact requires a deep understanding of the user's constraints. We learned how to optimize AI prompts to ensure the remedies suggested are truly low-cost and organic. Technically, we grew our skills in TypeScript architecture and database relational modeling.

What's next for FarmGuard AI

Community Knowledge Sharing: A feature to allow farmers to share successful organic remedy results with others in their region. Multi-language Support: Localizing the app into regional languages to reach the most underserved communities globally. Predictive Analytics: Using historical log data to warn farmers of upcoming seasonal pest outbreaks before they happen. Offline Support: Enabling the use of FarmGuard AI in Offline mode for farmers with bad connectivity

Built With

• html
• react
• supabase
• typescript
• vite