Krishi Sahayak AI

Multilingual AI assistant for Indian smallholder farmers. Hyperlocal weather, image-based pest/disease diagnosis (CNN), market price forecasting, alternative credit scoring, parametric crop insurance. Voice-first for low-literacy users.

By Aditya Singh Khichi, Full Stack Engineer, New Delhi, India.

Tech stack: Next.js, TypeScript, Python, Gemini 1.5 Pro, Groq, CNN, PWA.

Facebook Pragati AI for Impact: Top 25

Problem

Indian smallholder farmers face a stack of compounding problems: no timely hyperlocal weather forecasts, no quick pest/disease diagnosis, opaque mandi prices, financial exclusion from formal credit due to lack of land collateral, slow insurance payouts. Existing apps either solve one of these in isolation, or assume an English-literate user comfortable navigating a smartphone-first menu UI. Neither matches the actual user.

Approach

Built a unified platform organized around three pillars: Gyan Dhara (knowledge — AI weather + pest/disease doctor), Bazaar Bridge (market — price forecasting + verified input marketplace), Arthik Sahara (financial — alternative credit + parametric insurance). Hyperlocal weather advisories generated by tying ML forecasts to the farmer's specific crop and growth stage, delivered as multilingual voice + text. Pest/disease diagnosis via a CNN trained on Indian crop disease imagery with confidence scoring and IPM-prioritized treatment. Voice-first chatbot on top of Groq + Gemini 1.5 Pro so farmers can ask in Hindi ("कल मौसम कैसा रहेगा?") instead of typing English. Frontend Next.js PWA, backend Python serving ML models.

Outcome

Top 25 at Facebook's Pragati AI for Impact Hackathon. The platform shape — pillars over feature lists, voice over text, Hindi alongside English, parametric over traditional insurance — is the unlock. The lesson generalizes: when designing for users outside Silicon Valley defaults, the input modality is the product. A perfect app for an English-literate urban user is a useless app for a Hindi-speaking smallholder.

Live link: https://github.com/Raghav-45/pragati