Interviews Vector
Back to Roadmap
16
25 lessons

Multi-Agent & Swarms

Coordination, emergence, and collective intelligence.

01

Why Multi-Agent

Learn
TypeScript

One agent hits a wall. The smart move is not a bigger agent - it is more agents.

02

FIPA-ACL Heritage and Speech Acts

Learn
Python

Before MCP, before A2A, there was FIPA-ACL. In 2000 the IEEE Foundation for Intelligent Physical Agents ratified an agent communication language with twenty performatives, two c…

03

Communication Protocols

Build
TypeScript

Agents that can't speak the same language aren't a team. They're strangers shouting into the void.

04

The Multi-Agent Primitive Model

Learn
Python

Every multi-agent framework shipping in 2026 — AutoGen, LangGraph, CrewAI, OpenAI Agents SDK, Microsoft Agent Framework — is a point in a four-dimensional design space. Four pri…

05

Supervisor / Orchestrator-Worker Pattern

Build
Python

One lead agent plans and delegates; specialized workers execute in parallel contexts and report back. This is the pattern behind Anthropic's Research system (Claude Opus 4 as le…

06

Hierarchical Architecture and Decomposition Drift

Learn
Python

Hierarchical is supervisor nested. Manager agents over sub-managers over workers. CrewAI `Process.hierarchical` is the textbook version: a `manager_llm` dynamically delegates ta…

07

Society of Mind and Multi-Agent Debate

Build
Python

Minsky's 1986 premise — intelligence is a society of specialists — gets rediscovered every decade. In 2023 Du et al. turned it into a concrete algorithm: multiple LLM instances …

08

Role Specialization — Planner / Critic / Executor / Verifier

Build
Python

The most common multi-agent decomposition in 2026: one agent plans, one executes, one critiques or verifies. MetaGPT (arXiv:2308.00352) formalizes this as SOPs encoded into role…

09

Parallel Swarm and Networked Architectures

Build
Python

Contrast with supervisor: no central decider. Agents read a shared event bus, pick up work asynchronously, write results back. LangGraph explicitly supports "Swarm Architecture"…

10

Group Chat and Speaker Selection

Build
Python

AutoGen GroupChat and AG2 GroupChat share one conversation across N agents; a selector function (LLM, round-robin, or custom) picks who speaks next. This is the archetype of eme…

11

Handoffs and Routines (Stateless Orchestration)

Build
Python

OpenAI's Swarm (October 2024) distilled multi-agent orchestration to two primitives: **routines** (instructions + tools as a system prompt) and **handoffs** (a tool that returns…

12

A2A — The Agent-to-Agent Protocol

Build
Python

Google announced A2A in April 2025; by April 2026 the spec is at https://a2a-protocol.org/latest/specification/ and 150+ organizations back it. A2A is the horizontal complement …

13

Shared Memory and Blackboard Patterns

Build
Python

Two approaches coexist in 2026 multi-agent systems: the **message pool** (everyone sees everyone's messages, as in AutoGen GroupChat or MetaGPT) and the **blackboard with subscr…

14

Consensus and Byzantine Fault Tolerance

Build
Python

Classical distributed-systems BFT meets stochastic LLMs. In 2025-2026 three research directions emerged: **CP-WBFT** (arXiv:2511.10400) weighs each vote by a confidence probe; *…

15

Voting, Self-Consistency, and Debate Topology

Build
Python

The cheapest aggregation: sample N independent agents, majority-vote. Wang et al. 2022 self-consistency did this with one model sampled N times. Multi-agent extends it with **he…

16

Negotiation and Bargaining

Build
Python

Agents negotiate resources, prices, task allocations, and terms. The 2026 benchmark set is clear: NegotiationArena (arXiv:2402.05863) shows LLMs can improve payoffs ~20% via per…

17

Generative Agents and Emergent Simulation

Build
Python

Park et al. 2023 (UIST '23, arXiv:2304.03442) populated **Smallville**, a sandbox of 25 agents, with a three-part architecture: **memory stream** (natural-language log), **refle…

18

Theory of Mind and Emergent Coordination

Build
Python

Li et al. (arXiv:2310.10701) showed that LLM agents in a cooperative text game exhibit **emergent high-order Theory of Mind** (ToM) — reasoning about what another agent believes…

19

Swarm Optimization (PSO, ACO)

Build
Python

Bio-inspired optimization is making an LLM comeback. **LMPSO** (arXiv:2504.09247) uses PSO where each particle's velocity is a prompt and the LLM generates the next candidate; w…

20

MARL — MADDPG, QMIX, MAPPO

Learn
Python

The reinforcement-learning heritage of multi-agent coordination, which still informs LLM-agent systems in 2026. **MADDPG** (Lowe et al., NeurIPS 2017, arXiv:1706.02275) introduc…

21

Agent Economies, Token Incentives, Reputation

Learn
Python

Long-horizon autonomous agents (METR's 1-hour to 8-hour work-curve) need economic agency. The emerging **5-layer stack** is: **DePIN** (physical compute) → **Identity** (W3C DID…

22

Production Scaling — Queues, Checkpoints, Durability

Build
Python

Scaling multi-agent systems to thousands of concurrent runs requires **durable execution**. LangGraph's runtime writes a checkpoint after each super-step keyed by `thread_id` (P…

23

Failure Modes — MAST, Groupthink, Monoculture

Learn
Python

The reference taxonomy for 2026 is **MAST** (Cemri et al., NeurIPS 2025, arXiv:2503.13657), derived from 1642 execution traces across 7 state-of-the-art open-source MAS showing …

24

Evaluation and Coordination Benchmarks

Learn
Python

Five 2025-2026 benchmarks cover the multi-agent evaluation space. **MultiAgentBench / MARBLE** (ACL 2025, arXiv:2503.01935) evaluates star/chain/tree/graph topologies with miles…

25

Case Studies and 2026 State of the Art

Learn
Python

Three production-grade references to study end-to-end, each illustrating a different slice of multi-agent engineering. **Anthropic's Research system** (orchestrator-worker, 15x …