Add AI RAG Knowledge Assistant implementation

[Aman-Semwal]

This pull request introduces an AI-based Knowledge Assistant built using a Retrieval-Augmented Generation (RAG) architecture to enable intelligent question answering over a knowledge base.

The system processes user queries through a FastAPI backend, where the input question is converted into vector embeddings using Sentence Transformers. These embeddings are used to perform semantic similarity search in a FAISS vector database to retrieve the most relevant contextual information.

The retrieved context is then passed through a response generation pipeline to produce meaningful and context-aware answers for the user. A Streamlit frontend is integrated to provide an interactive interface where users can submit questions and view AI-generated responses in real time.

Key highlights of this implementation include:

• End-to-end RAG pipeline integrating retrieval and AI generation
• Semantic search using vector embeddings
• Efficient similarity search using FAISS vector database
• REST API backend built with FastAPI
• Interactive web interface built with Streamlit

This implementation demonstrates an Endee-style AI architecture that combines vector search with generative AI to build an intelligent knowledge retrieval system.