Building Multi-Agent Systems: From Concepts to Collaboration
summary
Shows a concrete MAS design that wires specialized agents (design, frontend, backend) into a planner + retrieval + generation + evaluator loop that actually builds drafts, searches the web for facts, refines outputs and runs auto-evaluation. Uses LangGraph for agent messaging and CrewAI for orchestration, gives each agent a local view and memory, warns that monolithic agents kill performance, and reports running iterative passes (commonly 3, sometimes 10–20) to reach production-quality results.
AI-written summary · about
chapters (8)
- 0:00Introduction to multi-agent systems
- 2:27How agents interact with tools
- 8:01Autonomy in multi-agent systems
- 10:57Agent communication networks
- 16:09Evolution of chatbots and agents
- 20:54Building an agent step by step
- 27:47Demo: Research agent in action
- 30:14Q&A session
related talks
26mHow to Build and Govern Production-Grade Multi-Agent Systems
45mAn Agent Walks Into a Bar... and Meets Another Agent: Multi-Agent Systems in Action
50mScalable Innovation: Exploring Multi-Agent Systems
28mThe Put That Sh*t On Agent: Building Multi-Agent AI Systems That Actually Talk to Each Other*
44mBuilding Agentic AI Systems: From Transactions To Conversations
27mMulti-Agent AI: Transforming Business Operations and Value Creation
01h 04mAgentic AI: Build a Multi-Agent Application with CrewAI
22mHOW TO BUILD YOUR OWN AI AGENT: THE COMPLETE ARCHITECTURE FOR ENGINEERING TEAMS