An AI-powered legal assistant that makes Pakistani laws accessible to everyone through Retrieval Augmented Generation (RAG), providing accurate legal information with proper citations.
An AI-powered legal assistant that makes Pakistani laws accessible to everyone through Retrieval Augmented Generation (RAG), providing accurate legal information with proper citations.
May 2025 - Jul 2025
AI/ML
Pakistani legal documents and laws were scattered across multiple sources and written in complex legal language that was difficult for the general public to understand and navigate. Citizens needed a way to quickly find relevant legal information and get clear, understandable explanations.
Building an AI system that could accurately retrieve and explain Pakistani legal information while maintaining proper citations and avoiding providing legal advice. The system needed to handle complex legal language and provide contextually relevant responses.
Developed a Retrieval Augmented Generation (RAG) system that processes legal documents from Pakistan Code, converts them into semantic vectors using OpenAI embeddings, stores them in a vector database, and uses Gemini Flush 2.0 for natural language understanding and response generation.