<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Software-Engineering on Alex Kim's blog</title><link>https://alex000kim.com/tags/software-engineering/</link><description>Recent content in Software-Engineering on Alex Kim's blog</description><generator>Hugo -- gohugo.io</generator><language>en-us</language><lastBuildDate>Sun, 26 Apr 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://alex000kim.com/tags/software-engineering/index.xml" rel="self" type="application/rss+xml"/><item><title>You've Been Doing Harness Engineering All Along</title><link>https://alex000kim.com/posts/2026-04-26-harness-engineering/</link><pubDate>Sun, 26 Apr 2026 00:00:00 +0000</pubDate><guid>https://alex000kim.com/posts/2026-04-26-harness-engineering/</guid><description>&lt;p>If you used LLMs for coding before the coding agents (~prior to Claude Code release in Feb last year), and if you got annoyed with typing the same context into LLM chat boxes, you&amp;rsquo;d start saving it into per-project markdown files. Many of those who used LLMs heavily ended up doing this eventually.&lt;/p>
&lt;p>Then we got Claude Code and other agents, and by mid-summer, this technique got formalized as the &lt;a href="https://agents.md/">&lt;code>AGENTS.md&lt;/code>&lt;/a> file and its tool-specific equivalents like &lt;code>CLAUDE.md&lt;/code>, &lt;code>.cursor/rules&lt;/code>, and &lt;code>.github/copilot-instructions.md&lt;/code>. A few of us have been doing this before it had a name.
At a bigger scale, that&amp;rsquo;s what happened to me with &amp;ldquo;Harness Engineering.&amp;rdquo;&lt;/p></description></item></channel></rss>