Agrisistance

🚀 Inspiration Agriculture is a vital part of many economies, especially in developing countries like Algeria — yet many farmers still rely on guesswork when choosing what crops to plant. This often leads to low yields, wasted resources, and financial risk. We were inspired to build AgriSistance to bridge this gap using technology — to empower farmers with smart, data-driven decisions powered by AI.

💡 What We Learned Through this project, we deepened our understanding of:

Working with geolocation and soil parameters in real-world datasets

Integrating AI in a way that’s simple and useful for non-technical users

Building full-stack web apps that are both scalable and user-friendly

The value of teamwork, especially when combining different skill sets (AI, UI/UX, backend, frontend, and business)

🛠️ How We Built It We started with UI design in Figma, then moved to building the frontend using React with TypeScript, and the backend with NestJS, Prisma, and PostgreSQL. We integrated OpenAI for the chatbot and developed custom logic for crop recommendation based on soil and location inputs. Real-time features were supported through a connected database, and we included an interactive map for easier land selection.

🧗 Challenges We Faced Finding reliable and clean datasets for soil and crop matching

Making the AI recommendations accurate and explainable

Ensuring the chatbot gives helpful, localized answers

Designing a clean, intuitive UI that would be usable by non-technical users (like farmers)

Despite these challenges, we delivered a working MVP that we’re proud of — and one that we believe has real-world potential.

Built With

• bolt
• figma
• nest
• postgresql
• react