July 19, 2026
A prompt-injection test passed at the text level and still failed at the system level. A practical way to prove an agent's trust boundaries hold.
Read more→→→In The Loop
In The Loop is my long-form publication on AI agents, context engineering, LLM systems, and the work behind useful AI. Expect technical guides, first-party experiments, failure modes, and practical decisions you can apply.
June 29, 2026
The hard half of context engineering is subtraction: deciding what to keep out of your AI's context, not just what to put in.
Read more→→→