Skip to content

/builds/12-factor-agentops

// Operations for AI agents

DevOps took a decade to figure out how to make infrastructure reliable, and SRE codified it. AI agents are the new chaos—and everyone is making the same mistakes we made with servers in 2010.

12-Factor AgentOps adapts those hard-won patterns for coding agents. For measuring human-AI collaboration sessions, see vibe-check.

12
factors
3
tiers
2+
years battle-tested

// the 12 factors

FOUNDATION
  • I. Automated Tracking
  • II. Context Loading
  • III. Focused Agents
  • IV. Continuous Validation
OPERATIONS
  • V. Measure Everything
  • VI. Resume Work
  • VII. Smart Routing
  • VIII. Human Validation
IMPROVEMENT
  • IX. Mine Patterns
  • X. Small Iterations
  • XI. Fail-Safe Checks
  • XII. Package Patterns

// key insight

Here's the translation table:

Context windows → load balancing
Token budgets → resource quotas
Validation gates → health checks
Agent routing → service mesh

// the engine

KNOWLEDGE FLYWHEEL

Every session produces learnings. ao forge extracts them, ao inject surfaces them next session. Knowledge compounds instead of decaying.

BROWNIAN RATCHET

Embrace variance (parallel agents), filter aggressively (validation gates), lock progress permanently (merge). Chaos in, quality out.

The deeper thesis: context orchestration, not agent orchestration →

The full methodology—all 12 factors with examples, failure patterns, and implementation guides.

12factoragentops.com →
GitHub →