KhetAI

KhetAI is a full-stack, offline-first web application designed to help Indian farmers diagnose crop diseases using AI.

Comment

Inspiration

In India, crop diseases cause massive losses for farmers every year. Many small-scale farmers don’t have access to expert agronomists or advanced diagnostic tools. We wanted to build something that empowers them directly - a simple, AI-powered solution that can detect crop diseases from images and provide actionable insights. The inspiration came from the idea of combining AI for social good with the urgent need for food security

What it does

KhetAI is an AI crop disease detector. Farmers or agricultural workers can upload or capture an image of a crop leaf, and the system analyzes it to identify potential diseases. It then provides a diagnosis along with suggested remedies or preventive measures. The goal is to make disease detection fast, accessible, and reliable for communities that need it most

How I built it

  • Used computer vision models trained on crop disease datasets.

  • Leveraged Python + TensorFlow/PyTorch for model development.

  • Built a lightweight web interface for easy image upload and results display.

  • Integrated with PostgreSQL for storing user queries and results.

  • Designed the UI/UX with Canva + Figma to keep it farmer-friendly

Challenges I ran into

  • Collecting and cleaning reliable datasets for crop diseases.

  • Ensuring the model works well with real-world, low-quality images taken by farmers.

  • Balancing accuracy vs. speed in predictions.

  • Deploying the model in a way that is lightweight and accessible even on low-end devices

Accomplishments that I'm proud of

  • Successfully built a working prototype that can detect multiple crop diseases.

  • Designed an interface that is simple enough for non-technical users.

  • Demonstrated how AI can be applied to solve community-driven problems.

  • Presented KhetAI at the Gemma 4 Good Hackathon, where it was recognized for impact

What I learned

  • The importance of data quality in AI projects.

  • How to design solutions with end-users in mind, especially those with limited tech literacy.

  • The challenges of deploying AI models in resource-constrained environments.

  • Teamwork and rapid prototyping under hackathon pressure.

What's next for KhetAI

  • Expanding the dataset to cover more crops and diseases.

  • Adding multilingual support for farmers across India.

  • Building a mobile app for offline use in rural areas.

  • Partnering with agriculture NGOs and government bodies to scale adoption.

  • Exploring integration with IoT sensors for real-time crop health monitoring

Built With

Share this project:

Updates