Abstract
<jats:p>Abstract - Legal research is increasingly challenged by the rapid growth of judicial data and the limitations of traditional keyword-based systems, which often fail to capture contextual meaning. LegalEase AI addresses this issue through an intelligent platform that leverages Retrieval-Augmented Generation (RAG) and Natural Language Processing (NLP) to enhance case discovery and reasoning. It enables users to input natural language queries and retrieves semantically relevant case laws while providing contextual summaries and predictive risk insights, simplifying research for practitioners and students. tive interface with search, analytics, and chatbot features. Results show improved accuracy and efficiency in case retrieval, with strong semantic relevance The system follows a structured pipeline including data ingestion, preprocessing, semantic embedding, retrieval, and AI-driven reasoning. Legal data is stored in PostgreSQL and converted into vector embeddings using SentenceTransformer models, which are indexed with FAISS for efficient similarity search. A FastAPI backend manages query processing and integrates LLaMA-based models for explanation and risk evaluation, while a Next.js Progressive Web Application provides an interac and consistent risk analysis. The system demonstrates how RAG and transformer models enable contextual understanding and scalable, user-friendly legal research solutions. Keywords: Retrieval-Augmented Generation (RAG), Legal AI, FAISS Vector Database, Natural Language Processing (NLP), Legal Case Retrieval, Risk Analysis, Semantic Search.</jats:p>