-
Landing page
SmartCropAI was inspired by the challenges faced by small farmers who lack access to real-time insights for crop health, irrigation, and profitability.
What it does is act as an agentic AI assistant that analyzes satellite imagery, soil data, weather forecasts, and market trends to provide simple, actionable advice on crop care, irrigation, and pest control through a mobile app or WhatsApp bot.
How we built it involved using CNNs for crop stress detection, LSTMs for weather and irrigation forecasting, and a decision engine powered by reinforcement learning to recommend actions, with IBM cloud resources enabling real-time predictions.
Challenges we ran into included integrating diverse datasets, ensuring accurate predictions with limited resources, and designing a farmer-friendly interface.
Accomplishments that we’re proud of include building a working prototype that can deliver daily/weekly recommendations and explain the reasoning behind them, making AI truly accessible to farmers. What we learned was not just the technical side of building predictive AI pipelines, but also the importance of user-first design and simplifying complex analytics into practical advice.
What’s next for SmartCropAI is expanding data coverage with higher-resolution imagery, adding voice-based support for local languages, and scaling the system to reach more farmers, helping them improve yield, sustainability, and resilience to climate change.
Log in or sign up for Devpost to join the conversation.