YojanaSetu
AI-powered GovTech platform automating document processing and eligibility verification for government schemes.
Next.jsReactTypeScriptPythonFastAPIOpenAI APIPostgreSQLAWS
The Problem
Citizens struggle to understand complex eligibility criteria for government schemes, leading to a 60% rejection rate due to incorrect documentation.
The Solution
Architected a RAG pipeline that processes unstructured government PDFs and provides an conversational interface for citizens to check eligibility in 12 languages.
Architecture
The frontend is built with Next.js (App Router) deployed on Vercel. User queries hit a FastAPI Python backend, which orchestrates calls to a Pinecone vector database containing chunked policy documents. The OpenAI API generates the final response, structured via LangChain.
Key Features
Automated PDF Parsing and OCR for Government Documents
RAG-based eligibility engine powered by LLMs
Multi-lingual Voice Agent for illiterate users
Secure citizen data vault with end-to-end encryption
Deep Dive
YojanaSetu is a revolutionary platform that leverages Large Language Models to bridge the gap between citizens and government schemes. It processes complex legal documents and government PDFs to instantly determine citizen eligibility with 94% accuracy. The platform features an AI agent that speaks 12 regional languages and has successfully reduced the average application processing time from 3 weeks to just 48 hours.
Future Roadmap
1Integrate WhatsApp Business API for direct citizen interaction
2Implement local, open-source LLMs (Llama 3) for data privacy
3Expand to state-level specific schemes