Not AI magic — software architecture. Structure belongs in code, not in prose.
Markus Hanses · 2026
A programming language whose instructions are only suggestions cannot scale. Reliability means moving logic out of prose and into the runtime.
When prose policy meets a language model — and the bill arrives in court.
Two voices on one website both speak for the company. Non-deterministic · weakly specified · hard to verify.
One request, one model. Under hope the output ships unchecked. Under contract, code checks every output and re‑prompts until it is valid — generate, verify, retry.
The minimal autonomous unit. The model itself decides whether and when to reach for a tool.
{…}
Database
Web search
One agent is enough for direct questions. It stops being enough when tasks need different reasoning, run in parallel, or a human must step in.
Several autonomous agents that jointly solve a task too complex for one — held together by three primitives.
Every control decision belongs in the architecture — not hidden in prompts.
A fixed, predefined order of steps — deterministic every run.
A hub delegates to specialists and decides the next step.
A coded graph; the LLM picks a branch, the structure routes.
Everyone reads and writes one growing, visible state.
Peers hand off to each other — no central conductor.
The system pauses at a modelled checkpoint and asks a human.
Picking a pattern is choosing how much agency you delegate. Left trades capability for control; right trades control for capability.
Spectrum framing follows Anthropic (Building Effective Agents) & LangChain. Placement is editorial.
A single call defends one boundary — one prompt in, one answer out. A multi-agent system must defend as many boundaries as it has agents, tools, and hand-offs. Every inter-agent message is a trust boundary: one compromise cascades to every peer that trusts it.
Catalogued by OWASP (LLM Top 10 · 17 Agentic Threats) and threat-modelled with MAESTRO. Treat every hand-off as untrusted — re-verify, don’t trust by origin.
Multi-agent systems are software architecture: specialised roles, explicit state, controlled flow.
Explore the patterns, frameworks & runnable demos in the knowledge base →