Skip to content

One reliability loop, applied at every layer.

// the operating loop

The stack changes. Observe, automate, validate, recover, and hand off stay.

DEPENDENCIESUnderstand what the system actually rests on, from network and storage through software and people.
STATEMake desired state, ownership, and change history explicit.
OBSERVEInstrument the real system so drift and failure become visible.
AUTOMATETurn the repeatable path into a paved road with rollback points.
VALIDATEUse evidence and an independent check before promoting change.
RECOVERDesign the failure path before the system needs it.
HAND OFFLeave operators able to run and improve the system without its author beside them.

// operator beliefs

Seven from running the platforms. The discipline came first; the beliefs are what survived it.

01 · AI IS A SLOP CANNON

Raw output is cheap, infinite, and wrong somewhere. Validation, provenance, and release discipline forge slop into trust.

RECEIPT

The AgentOps repo runs the loop on itself: a fresh verdict records what changed and which checks support it.

02 · ZERO-TRUST FOR AI OUTPUT

Treat model output like untrusted network traffic: validation gates before execution, always. The agent that wrote the code never grades its own work.

RECEIPT

AgentOps binds a fresh verdict to unchanged acceptance, the actual changed files, and evidence for each acceptance check.

03 · AIR-GAPPED IS HARDER

Air-gapped delivery removes the easy fallback. Constraints manufacture reliability.

RECEIPT

Self-hosted models served on a 100+ GPU fleet under air-gapped operational constraints.

04 · GIT IS INSTITUTIONAL MEMORY

Commits record what changed. Beads record why. A maintained wiki preserves the reviewed patterns worth reusing.

RECEIPT

AgentOps keeps intent snapshots, candidate manifests, evidence references, and fresh verdicts as ordinary inspectable files.

05 · CONTEXT IS A BUDGET

Always-on instructions compete with the task itself. Keep universal rules small and load procedures or references only when the work needs them.

RECEIPT

AgentOps skills carry procedures and references on demand; the active bead stays the durable source of intent.

06 · RUN THE PLATFORM LIKE A FAB

Maximum throughput at maximum yield, gates at every step. Tokens, deploys, decisions: same discipline.

RECEIPT

GitOps promotion across 100+ clusters: 58 applications, automated rollback, full artifact traceability.

07 · EVIDENCE OVER NARRATIVE

Every claim traces to an artifact. If it can't be shown, it isn't said.

RECEIPT

Every number on this site maps to one receipts inventory; /work states each case as problem, constraint, built, number.

// teacher beliefs

Five from teaching it: flag officers first, then engineers, now everyone else.

08 · PLAIN WORDS ARE A TECHNICAL SKILL

If it can't be explained simply, it isn't understood yet. Briefing a flag officer and teaching an engineer are the same discipline: respect the listener, kill the jargon, keep the truth.

RECEIPT

Three years of daily technical briefings to flag officers, SES, and O-6 leadership.

09 · PRIMITIVES FIRST, NO MAGIC

Frameworks hide the thing you'll be debugging at 2 a.m. Teach the raw parts; let people earn their abstractions.

RECEIPT

/training runs live in a terminal, no slides, and every session ends with a runnable artifact the team keeps.

10 · THE MULTIPLIER ETHOS

The metric that matters: everyone near you ships faster. Developers, and now AI users, the new developers.

RECEIPT

Onboarding docs that cut engineer ramp-up from ~3 months to ~1 week.

11 · THE MISSION

Models make capability cheap. The job is getting dependable AI into constrained environments without burning the house down.

RECEIPT

/ai-partner translates the same production discipline into plain language for people who will never read a runbook.

12 · SAME BOTTLENECK, NEW WORKFORCE

DevOps made software delivery reviewable and recoverable. Agent delivery still ships on hope unless the same gates exist.

RECEIPT

12-Factor AgentOps: the DevOps playbook rewritten for agent delivery, public at 12factoragentops.com.

// see it run

Real Claude Code and Codex sessions running the AgentOps loop end to end. Nineteen seconds, silent, lightly edited for pace.

// one public receipt

The beliefs, operationalized as inspectable artifacts.

AgentOps

An operating loop for coding agents: one intent, one bounded build, one fresh verdict, plus a Go CLI, portable skills, and marketplace kits · 400+ stars on GitHub. This site, the CLI, and the kits were built with the workflow they document.

// independent practice

I also run this loop with engineering teams on coding-agent workflows. Start at /training.