GiovanniPasq/agentic-rag-for-dummies
A modular framework for building advanced Agentic RAG systems using LangGraph, featuring intelligent query processing, conversation memory, and human-in-the-loop clarification.
Core Features
Detailed Introduction
This project provides a modular and extensible architecture for building Agentic RAG (Retrieval-Augmented Generation) systems with LangGraph. It addresses the common gap in RAG tutorials by offering both learning materials and a robust framework. Key features include hierarchical indexing for optimal document retrieval, conversation memory for natural dialogue, and an intelligent four-stage query processing workflow with human-in-the-loop clarification and self-correction. It supports various LLM providers and allows for easy component swapping, making it ideal for both learning and developing sophisticated RAG applications.