<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Ownership on Armstrong Yan</title><link>https://yanqian.github.io/tags/ownership/</link><description>Recent content in Ownership on Armstrong Yan</description><generator>Hugo</generator><language>en</language><lastBuildDate>Sat, 23 May 2026 20:33:47 +0800</lastBuildDate><atom:link href="https://yanqian.github.io/tags/ownership/index.xml" rel="self" type="application/rss+xml"/><item><title>AI-Native Software Engineering, Part 4: Human Judgment Against Vibe Coding</title><link>https://yanqian.github.io/posts/publish/against-vibe-coding-why-human-judgment-still-matters/</link><pubDate>Sat, 23 May 2026 20:33:47 +0800</pubDate><guid>https://yanqian.github.io/posts/publish/against-vibe-coding-why-human-judgment-still-matters/</guid><description>&lt;p>You can automate implementation. You can automate evaluation. But judgment remains stubbornly human.&lt;/p>
&lt;p>This is Part 4 of the AI-Native Software Engineering series.&lt;/p>
&lt;p>It continues from &lt;a href="https://yanqian.github.io/posts/publish/software-is-becoming-search-why-engineers-are-turning-into-constraint-designers/" >AI-Native Software Engineering, Part 3: Software as Search&lt;/a>.&lt;/p>
&lt;p>The first article asked how understanding forms.&lt;/p>
&lt;p>The second asked how correctness forms.&lt;/p>
&lt;p>The third argued that software is starting to look like search.&lt;/p>
&lt;p>This article pushes back on a dangerous misunderstanding:&lt;/p>
&lt;div class="highlight">&lt;pre tabindex="0" class="chroma">&lt;code class="language-text" data-lang="text">&lt;span class="line">&lt;span class="cl">If we have agents, harnesses, and constraints,
&lt;/span>&lt;/span>&lt;span class="line">&lt;span class="cl">software can produce itself.
&lt;/span>&lt;/span>&lt;/code>&lt;/pre>&lt;/div>&lt;p>That is not true.&lt;/p></description></item></channel></rss>