Build an End-To-End Modular RAG LangGraph AI Agent with Web Search | PDF Upload, Pinecone, Groq

🚀 Build your own intelligent AI Agent that combines custom PDF-based knowledge with real-time web search using Retrieval-Augmented Generation (RAG), LangGraph, and Groq's LLaMA 3 model!

🔍 This video showcases a full-stack, modular AI system that:
- Accepts PDF uploads and stores content in Pinecone.
- Uses LangGraph for agent orchestration (Router, RAG, Web Search, Answer)
- Offers a user toggle to control web search behavior
- Displays transparent AI decisions with full traceability
- Leverages Groq for fast LLM inference + Tavily for web search

🧠 Technologies Used:
- FastAPI (Backend)
- Streamlit (Frontend UI)
- LangGraph + LangChain
- Groq LLaMA 3
- Pinecone Vector DB
- HuggingFace Sentence Transformers (MiniLM)
- Tavily Search API

📁 PDF Download & Code: https://github.com/snsupratim/agentbot

Chapters :
00:00 ---- Intro/Demo
07:08 ---- Concept Explanation
15:20 ---- Prerequisite
15:41 ---- Setup & Installation
17:56 ---- Test FastAPI in Postman
21:13 ---- Backend Programming
02:35:54 --- Test Backend in Postman
02:44:06 --- Connect with Streamlit UI
02:47:00 --- Final Run
💬 Drop your questions in the comments if you want help building your own version!

#langchain #AI #LangGraph #Groq #RAG #FastAPI #Streamlit #Python #Pinecone #Tavily

Видео Build an End-To-End Modular RAG LangGraph AI Agent with Web Search | PDF Upload, Pinecone, Groq канала sn dev