AI Research Assistant: A Multi-Document Agentic RAG System

Published:

This project is an advanced web application that acts as an AI research assistant, capable of reading, comparing, and synthesizing information from my two NLP research papers. It moves beyond a simple Q&A bot by using a LangChain Agent that can autonomously decide which document to consult to answer complex, comparative, and conversational questions.

The core of this project is a ReAct (Reasoning and Acting) agent that can use multiple tools—one for each research paper—to gather information before formulating a final, synthesized answer. This showcases a modern, sophisticated approach to building AI systems that can handle more dynamic and complex user queries.