Fuses behavioral and error telemetry to catch three classes of production issues, investigates root cause on an Aurora-adapted agent, patches and previews the fix, then closes the loop through a human-gated canary release with automatic rollback — the one part of this pipeline nobody has published closed end to end.
The earlier “layers” mixed together tools, pipeline stages, and infrastructure. This map separates the three jobs AcesoLoop performs during one incident.
Follow the solid arrows left to right. Observe finds the problem, investigate proves the cause and prepares a fix, then release safely decides whether the fix stays or rolls back.
Combine what users did with what the system reported, then decide whether the change is abnormal.
Funnels + session replay
Errors + stack traces
Synthetic user sessions
One comparable signal shape for every source.
Classifies crashes, slowdowns, or silent flow breakage.
Use Aurora to investigate, then make the patch earn its way through tests and a real preview.
Gather evidence, rank hypotheses, explain the root cause.
Generate a scoped code change with an open-weight model.
Reject and retry if the existing tests fail.
Replay synthetic traffic and confirm the original anomaly clears.
A person approves the evidence; a small canary proves the release before full promotion.
Review diagnosis, diff, and preview evidence.
Send a weighted slice of traffic to the new version.
Monitor recovery and catch any new anomaly.
The map above shows the three jobs. These nine states are the exact Postgres-backed state machine inside those jobs.
A reasoned estimate, weighted by effort share per stage — not a measured number, since this combination hasn't been built before.