<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Artificial Intelligence - Joe Hutchinson</title><link>https://www.joehutch.com/categories/artificial-intelligence/</link><description>Recent content in Artificial Intelligence on Joe Hutchinson</description><generator>Hugo -- gohugo.io</generator><language>en-US</language><lastBuildDate>Sun, 17 May 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://www.joehutch.com/categories/artificial-intelligence/index.xml" rel="self" type="application/rss+xml"/><item><title>AI Agents Don’t Think in Shell Sessions</title><link>https://www.joehutch.com/posts/agents-and-toolchain-managers/</link><pubDate>Sun, 17 May 2026 00:00:00 +0000</pubDate><guid>https://www.joehutch.com/posts/agents-and-toolchain-managers/</guid><description>AI coding agents expose a subtle flaw in many modern development environments: tools designed around long-lived human shell sessions often break down when commands are executed through short-lived, non-interactive subprocesses. Devbox, worktrees, cloud agent environments, and shell activation all start behaving differently once the “developer” issuing commands is no longer human.</description></item></channel></rss>