Factory 2.0: from coding agents to software factories

Thesis: individual coding speedups aren’t enough. Org-wide productivity needs an interconnected agent-native system that observes itself—the software factory. Loop: external signals → triage → plan → build → test → review → secure → ship → monitor → new signals.

Three requirements: model independence (route by cost/perf/speed), sovereign intelligence (learning stays inside your walls), continual learning across stages so security findings inform review and incidents correlate to PRs.

Autonomy as maturation spectrum: skills → automations → remote persistence → multi-day Missions. Engineers own governance and outcomes, not sole authorship of every line.

Why it matters for code review

Positions review as a first-class instrumented station in a closed loop—not a bolt-on chatbot—sharing context with security, QA, and incident response.