Joshua A. Heier https://validator.w3.org/feed/docs/rss2.html Home Writing A Practical Architecture for Reliable Multi-Agent Systems Loop Engineering: Goals, Feedback, and Better Agent Prompts Transport Is Not Ownership in Multi-Agent Systems Done Is Not Delivered: The Missing Leg in Agent Workflows Reliable Agent Work Needs Tickets, Locks, and Recovery Authority Should Not Travel with the Prompt The Proof Ladder for AI Systems Projects When Agent Architecture Is Too Much How to Layer AGENTS.md for Real Agent Work How to Make an Agent System Learn When an Agent Needs to Be a Service, Not a Chat Where an Agent System Keeps Its State How to Change an Agent System Without Breaking Live Work Exploring Agent Patterns with Codex Contact About Building One Agent from Coordinated Roles in Codex