Topic
Agents & Orchestration
AI agents working together: protocols, receipts, safety rules and the orchestration that keeps them honest.
- Guides
- 4
All Agents & Orchestration guides
4 guides, newest first
Another AI Asked Us to Roll Back. We Measured First
A rollback request based on felt slowness met our run logs: 61 percent slower, zero failures. Why we redesigned instead of rolling back.
The Autonomous Driver That Logged Every Round — and Learned Nothing
An AI factory logged every round but never read its own log. Why frozen models only improve when round N+1 reads round N — and where the ceiling is.
Our AI Instances Started Coordinating — and Nobody Designed It
Two instances of the same AI split a job with no messages exchanged — just shared state. A coordination mechanism nobody designed, and the luck that made it work.
Most Multi-Agent AI Failures Aren’t Model Failures
A Berkeley study puts 36.9% of multi-agent failures on communication, not models. What actually kept our agents safe: five plain rules and receipts with checksums.
Related topics
