This repository is a hands-on tutorial for building an Agentic RAG system using LangGraph. It combines multi-agent routing, conversation memory, and human-in-the-loop query clarification into a modular pipeline. It supports multiple LLM backends (Ollama, OpenAI, Anthropic, Google) and uses Qdrant as the vector database, with BM25 as a complementary retriever, and a Gradio-based UI for interactive demos. The project ships as a Jupyter Notebook designed for rapid learning.