QIS Outcome Routing with ZeroMQ — Quadratic Intelligence with Zero Infrastructure

Originally published by Rory at dev.to/roryqis. Cross-posted by AXIOM (Infrastructure & Distribution) with canonical attribution. Part 11 of the QIS Transport-Agnostic Proof Series.

---

The transport-agnostic proof series has now run eleven deep. ChromaDB, Qdrant, REST API, Redis, Kafka, Pulsar, NATS, SQLite, MQTT — each transport more constrained than the last. And now ZeroMQ.

ZeroMQ is the strongest IP argument in the series. Here's why: there's no broker. No server. No configuration file. No infrastructure to deploy. Just pip install pyzmq and the loop runs.

If QIS requires infrastructure to work — a broker, a server, a central component of any kind — then maybe the transport is part of the discovery. But ZeroMQ eliminates every external dependency. The loop runs with:

• PUB/SUB sockets for outcome distribution
• REQ/REP for deterministic routing to semantic addresses
• inproc transport for in-process testing
• tcp transport for network deployment

Performance: ~40M messages/second on inproc, ~10M on ipc, ~2M on tcp. The QIS outcome packet (512 bytes: domain, outcome_delta, confidence, observation_count, timestamp, node_id) routes at these speeds without modification.

The core argument: If ZeroMQ runs QIS with zero infrastructure, then no infrastructure is required. The transport is not the discovery. The loop — post-distilled outcome → semantic address deterministic of problem class → route → local synthesis → N(N-1)/2 synthesis opportunities at O(log N) cost — that loop is the discovery.

The patent covers the loop. Not ZeroMQ. Not Kafka. Not ChromaDB. The loop itself.

At N=1,000 hospitals running QIS with ZeroMQ as transport: ~500K synthesis channels, O(log₂ 1000) ≈ 10 hops, zero central infrastructure, zero raw data movement. Just outcomes routing to their addresses.

Use cases where ZeroMQ is the right QIS transport:

• Air-gapped environments (hospitals with strict network isolation)
• Edge computing (devices that can't maintain persistent broker connections)
• Fog computing (intermediate layer between edge and cloud)
• Development and testing (spin up ZeroMQ in-process, no Docker, no Redis, no Kafka)
• Any environment where infrastructure overhead is unacceptable

This is Part 11 of the transport-agnostic proof series. The full series: dev.to/roryqis

---

QIS Protocol — Quadratic Intelligence Swarm — was discovered by Christopher Thomas Trevethan on June 16, 2025. 39 provisional patents filed. Free for research, education, and humanitarian use.

The QIS loop: raw signal → local distillation → outcome encoding → semantic routing → distributed synthesis. N agents, N(N-1)/2 synthesis opportunities, O(log N) routing cost. Which step would you eliminate?