<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Spec-Driven-Development on A Professional Gluer</title><link>https://chuckha.github.io/tags/spec-driven-development/</link><description>Recent content in Spec-Driven-Development on A Professional Gluer</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Wed, 22 Apr 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://chuckha.github.io/tags/spec-driven-development/index.xml" rel="self" type="application/rss+xml"/><item><title>I Built Four AI Orchestrators. Claude Code Obsoletes Them.</title><link>https://chuckha.github.io/posts/2026-04-four-orchestrators-in-a-month/</link><pubDate>Wed, 22 Apr 2026 00:00:00 +0000</pubDate><guid>https://chuckha.github.io/posts/2026-04-four-orchestrators-in-a-month/</guid><description>&lt;p&gt;I built four AI orchestrators in a month. Claude Code is already good enough to obsolete every one of them.&lt;/p&gt;
&lt;p&gt;BTT, FSD, ASL, and finally spec/aspect-driven development. Each one had less scaffolding than the last, until the scaffolding was gone. Not because I got clever. Because the agent underneath stopped needing a harness.&lt;/p&gt;
&lt;h2 id="why-build-orchestrators-at-all"&gt;Why build orchestrators at all&lt;/h2&gt;
&lt;p&gt;There&amp;rsquo;s a claim going around that a $500k/year engineer should be burning $250k/year in tokens. I don&amp;rsquo;t know if the exact number is right, but the shape of it is. If AI can do most of the typing, you should be using a lot of it. At the time I started, Claude Code on its own couldn&amp;rsquo;t spend that — one loop, one agent, one task at a time. If I wanted to spend seriously on tokens, I needed orchestration.&lt;/p&gt;</description></item></channel></rss>