Phase 16

Multi-Agent and Swarms

Phase 16: Multi-Agent and Swarms. 25 hands-on lessons building AI from first principles in the browser. Free reading; graded exercises and certificate with lifetime access.

  1. Why Multi-Agent?
  2. Heritage of FIPA-ACL and Speech Acts
  3. Communication Protocols
  4. The Multi-Agent Primitive Model
  5. Supervisor / Orchestrator-Worker Pattern
  6. Hierarchical Architecture and Its Failure Mode
  7. Society of Mind and Multi-Agent Debate
  8. Role Specialization — Planner, Critic, Executor, Verifier
  9. Parallel / Swarm / Networked Architectures
  10. Group Chat and Speaker Selection
  11. Handoffs and Routines — Stateless Orchestration
  12. A2A — The Agent-to-Agent Protocol
  13. Shared Memory and Blackboard Patterns (graded)
  14. Consensus and Byzantine Fault Tolerance for Agents (graded)
  15. Voting, Self-Consistency, and Debate Topology (graded)
  16. Negotiation and Bargaining (graded)
  17. Generative Agents and Emergent Simulation
  18. Theory of Mind and Emergent Coordination
  19. Swarm Optimization for LLMs (PSO, ACO) (graded)
  20. MARL — MADDPG, QMIX, MAPPO
  21. Agent Economies, Token Incentives, Reputation (graded)
  22. Production Scaling — Queues, Checkpoints, Durability
  23. Failure Modes — MAST, Groupthink, Monoculture, Cascading Errors
  24. Evaluation and Coordination Benchmarks
  25. Case Studies and the 2026 State of the Art
0 lifetime access. Curriculum based on AI Engineering from Scratch by Rohit Ghumare (MIT, used under attribution).