Mar 25
Loading video player...
In this video, we build a real AI agent using LangGraph, going beyond simple prompts to design a planning-based, production-style agent system that plans before execution, decides when internet research is required, breaks tasks into parallel subtasks, uses multiple worker agents, adds citations and images automatically, and generates a complete blog end-to-end; you’ll learn how modern AI agents are actually designed using the orchestrator–worker architecture, along with key concepts like router, planner, worker, and reducer nodes, research-powered agents with Tavily, image-aware content generation, and how to turn an AI agent into a real, interview-ready project using LangGraph, LangChain, OpenAI, Gemini, Python, and Streamlit. Resources: https://github.com/campusx-official/blog-writing-agent Gen AI Courses: https://learnwith.campusx.in/s/store/courses/Paid%20Courses DSA Course: https://learnwith.campusx.in/courses/DSA-69527ab734c0815fe15a08d9 📱 Grow with us: CampusX' LinkedIn: https://www.linkedin.com/company/campusx-official CampusX on Instagram for daily tips: https://www.instagram.com/campusx.official My LinkedIn: https://www.linkedin.com/in/nitish-singh-03412789 Discord: https://discord.gg/PsWu8R87Z8 E-mail us at support@campusx.in ⌚Chapters⌚ 00:00 – Introduction & What We’re Building 02:10 – What Is a Planning AI Agent (vs Direct Execution) 06:30 – Demo: AI Blog Writing Agent in Action 11:00 – High-Level Agent Architecture Overview 15:30 – Orchestrator–Worker Pattern Explained 20:45 – Building the Basic Blog Writing Agent (Stage 1) 32:00 – Improving Output with Better Planning & Prompts 36:30 – Adding Internet Research with Router & Tavily (Stage 2) 49:50 – Research-Aware Planning & Worker Execution 59:40 – Adding Automatic Image Generation (Stage 3) 01:14:20 – Reducer Logic: Merging Content & Placing Images 01:16:20 – Final Blog Output with Images 01:16:30 – Building the Streamlit UI (Stage 4) 01:18:40 – Project Wrap-up & Portfolio Advice