Deep-dive research briefings on specific topics, newest first.
The emerging discipline of supervisory engineering in agent-assisted workflows
Teams adopting coding agents need to understand how engineering roles, skills, and team structures are shifting
Comparing TDAD, Arbiter, and existing approaches to prompt regression testing for AI agents
Every production agent team eventually discovers that small prompt changes cause silent behavioral regressions — this scout maps the landscape of available detection and prevention approaches
Autoresearch — the autonomous experiment loop pattern for iterative optimization
Directly applicable to automated optimization, quality improvement, and security hardening workflows in agent infrastructure
Model Context Protocol production deployment issues — what breaks when MCP leaves the demo and hits real infrastructure
MCP is becoming the de facto agent-to-tool standard; understanding its production gaps is essential before committing to it as architectural foundation
