Hands-on with AI agents: From scratch to LangGraph
summary
Builds a pure Python React agent that runs thoughts → actions → observations using the OpenAI API (GPT‑4) plus function calling and stubbed tools for weather, flights and math; the demo creates a travel-booking prototype (supervisor → flight advisor) that works for simple queries but sometimes fails and needs loop limits (five turns) and better tool specs. Then ports the same logic into LangGraph stateful graph to add persisted state, human‑in‑loop checks and multi‑agent wiring.
AI-written summary · about
chapters (6)
- 0:00AI agents introduction
- 8:07React agents workflow
- 18:47Function calling in agents
- 26:28Building React agent in Python
- 45:57LangGraph for multi-agent systems
- 50:13LangGraph state nodes and edges
related talks
01h 27mBlazing the AI Trail: Using LangGraph to Conquer the Oregon Trail
01h 59mBuilding Agentic and Multi-Agent Systems with LangGraph
25mFrom Prompt to Production: Building Autonomous AI Agents in Node.js
29mDemystifying AI Agents with Python Code
02h 08mIntroduction to AI Agents
40mAI Agents in Practice
41mGoing beyond copilot with AI Agents
48mAI Coding Agents and how to code them