Build the mental model

A multi-agent system is a team of model-driven workers with explicit roles and communication. More agents do not create free intelligence; they add decomposition, parallelism, and independent checks while also adding coordination cost and more places to fail. The boundary matters: An agent combines a model with state, tools, and a loop; a workflow can be deterministic; multi-agent means multiple decision-making loops. Calling the same model three times is not meaningful specialization, and an agent conversation is not a substitute for shared transactional state.

Draw the system in prose as four boxes: need → evidence → decision → consequence. The need belongs to a person or workflow, not the model. Evidence is what the system can legitimately inspect. The decision is the transformation multi-agent systems performs. The consequence is what changes for a user, operator, or downstream system. If you cannot fill every box, the design is still a label rather than a working mental model.

For a software-release workflow, ask what happens when the model is absent. That baseline reveals the actual value. Then ask what remains deterministic: identity, permissions, arithmetic, record updates, and irreversible actions should not become fuzzy merely because a model participates. The model can propose or interpret; an application still owns policy and state.

An agent combines a model with state, tools, and a loop; a workflow can be deterministic; multi-agent means multiple decision-making loops. Calling the same model three times is not meaningful specialization, and an agent conversation is not a substitute for shared transactional state. This distinction is practical. It tells you where to inspect a failure and which component can repair it. Avoid explaining the concept as “the AI understands everything.” Name the artifact moving between stages—a token sequence, retrieved passage, ranked candidate, stored memory, trace, image latent, or agent message.

A good explanation also includes uncertainty. Inputs may be incomplete, learned behavior is probabilistic, and proxies can disagree with real outcomes. That does not make the system unusable; it means the workflow needs a fallback and a way to expose uncertainty rather than hiding it in fluent prose.

Use this compact record:

| Question | Concrete answer to supply | |---|---| | User job | One action the person is trying to complete | | Input boundary | Data allowed into the system | | Model contribution | The uncertain judgment or generation | | Deterministic guard | Rule, permission, schema, or calculation | | Success signal | Observable outcome, split by important group | | Escape hatch | Retry, fallback, escalation, or stop |

For this topic, a plausible first signal is not “the output looks intelligent.” It is a task outcome tied to the concept and checked on representative cases. Save this sketch; later pages add controls and measurements without changing the user job.

Teach the concept using the analogy above, then deliberately state where the analogy breaks. The kitchen, notebook, workbench, team, or librarian metaphor omits numerical limits and operational ownership. A learner has mastery when they can leave the metaphor and describe the actual information flow.