kisan ai

Inspiration

Farmers are the backbone of our economy, yet many still struggle with limited access to timely, understandable, and localized agricultural information. Most digital solutions are text-heavy, English-centric, and not designed for real-world farming conditions.

We were inspired to bridge this gap by building a voice-first, AI-powered assistant that farmers can interact with naturally—just by speaking or sharing images—making advanced agricultural intelligence accessible to everyone, regardless of literacy or language.

🚀 What it does

Kisan Mitra AI is an intelligent farming assistant that helps farmers make better decisions throughout the crop lifecycle.

🎤 Voice Assistant: Farmers can ask questions in their local language and get instant answers 📸 Crop Quality Grading: Upload images to assess produce quality and estimate market price 🌱 Soil Analysis: Extract and analyze soil health data from images 🌾 Crop Advisory: Get recommendations on what to plant, when, and how 📊 Real-time Insights: Access live mandi prices and weather updates

It acts like a personal agricultural advisor in your pocket.

🛠️ How we built it

We built Kisan Mitra AI using a scalable, cloud-native architecture:

Frontend: React + TypeScript for a responsive and interactive UI Backend: .NET 8 Web API with clean architecture Cloud (AWS): Transcribe → Speech-to-text Bedrock (Claude) → AI reasoning Rekognition → Image analysis Textract → OCR for soil cards Polly → Text-to-speech DynamoDB & S3 → Data storage Integration: REST APIs for weather and market data

The system processes voice, image, and text inputs in real-time to deliver personalized insights.

⚠️ Challenges we ran into 🌐 Multilingual voice handling: Ensuring accurate speech recognition across Indian languages 📸 Image variability: Handling inconsistent lighting and quality in crop images ⚡ Real-time performance: Reducing latency in AI responses for smooth user experience 🔗 Service integration: Seamlessly orchestrating multiple AWS AI services 📊 Data reliability: Ensuring accurate and up-to-date mandi price data 🏆 Accomplishments that we're proud of ✅ Built a fully functional end-to-end AI system ✅ Implemented voice-first interaction, making it accessible to non-technical users ✅ Integrated multiple AI capabilities (voice, vision, NLP) into one platform ✅ Designed a scalable serverless architecture on AWS ✅ Created a solution with real-world impact potential for farmers 📚 What we learned How to design and deploy AI-powered, serverless applications at scale Practical experience integrating multiple AWS AI services Importance of user-centric design, especially for rural and non-technical users Handling real-world data challenges like noise, inconsistency, and variability Building modular and maintainable architectures using clean design principles 🔮 What's next for Kisan Mitra AI 🌍 Expand support to 10+ Indian regional languages 📱 Launch mobile apps (Android & iOS) 📡 Enable offline mode for low-connectivity rural areas 🌦️ Add weather-based alerts and predictive insights 🦠 Implement crop disease detection using advanced vision AI 🏪 Build a farmer-to-marketplace connection platform 🤝 Integrate with government schemes and subsidies

Built With

• amazon-web-services
• amazonbedrock
• amazoncognito
• amazonpolly
• amazonrekognition
• amazons3
• amazontextract
• amazontimestream
• amazontranscribe
• aspnetcore
• awscli
• awslambda
• awsstepfunctions
• cleanarchitecture
• corewcf
• csharp
• dotnet8
• dynamodb
• fscheck
• git
• github
• javascript
• materialui
• react
• reduxtoolkit
• restapi
• soap
• sql
• tailwindcss
• typescript
• visual-studio
• vscode
• webaudioapi
• xunit