AgroAdvisor Ghana

What Inspired Me

Ghana has over 2 million smallholder farmers who grow the food that feeds the nation. Yet most of them have no access to agricultural extension officers — the experts who advise on crop diseases, watering schedules, and harvest timing. I grew up seeing this gap firsthand. A farmer loses an entire tomato harvest because they didn't know the yellowing leaves were a sign of nitrogen deficiency — not a virus.

I wanted to put expert farming knowledge in every farmer's pocket, for free, in plain simple English, available anytime on any phone.

How I Built It

I built AgroAdvisor Ghana entirely using MeDo — Baidu's AI-powered no-code app builder. Here's how the process went:

1. I described the app I wanted to MeDo in plain natural language
2. MeDo generated a full requirements document covering all pages, features, and logic
3. I reviewed and confirmed the requirements, then MeDo built the entire full-stack app
4. I iterated using MeDo's multi-turn chat to refine the UI, AI persona, and responses
5. I deployed the app with one click to a public URL

The most impressive feature MeDo generated was the AI extension officer persona — the app responds to farming problems like a real Ghanaian agricultural expert, giving diagnosis, step-by-step advice, watering schedules, pest warnings, and harvest timing, all tailored to the farmer's specific crop, region, and season.

What I Learned

• How to turn a real-world problem into a working app using AI-powered no-code tools
• How to structure conversations with MeDo to get precise, production-quality results
• That you don't need to write a single line of code to build something meaningful
• The power of context — giving the AI a specific persona (Ghanaian extension officer) dramatically improved the quality and relevance of responses

Challenges I Faced

• Getting the AI to respond in truly simple English suitable for farmers with limited tech experience required careful prompt iteration in MeDo
• Designing a form that felt intuitive for non-tech-savvy users on mobile phones
• Scoping the app realistically — I had to leave out features like image upload and weather API integration for this version and focus on delivering core value cleanly

What's Next

• Add image upload so farmers can photograph sick crops for visual diagnosis
• Integrate real-time weather data by region
• Add Twi and other Ghanaian language support
• Partner with NGOs and Ghana's Ministry of Food and Agriculture for wider reach

Built With

• ai
• baidu
• css
• ernie
• html
• javascript
• medo
• no-code
• rest