<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Compilers on MacWorks</title><link>https://macworks.dev/tags/compilers/</link><description>Recent content in Compilers on MacWorks</description><generator>Hugo</generator><language>en</language><atom:link href="https://macworks.dev/tags/compilers/index.xml" rel="self" type="application/rss+xml"/><item><title>Hacker News</title><link>https://macworks.dev/docs/today/hackernews-2026-05-30/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://macworks.dev/docs/today/hackernews-2026-05-30/</guid><description>&lt;h1 id="hacker-news--2026-05-30"&gt;Hacker News — 2026-05-30&lt;a class="anchor" href="#hacker-news--2026-05-30"&gt;#&lt;/a&gt;&lt;/h1&gt;
&lt;h2 id="top-story"&gt;Top Story&lt;a class="anchor" href="#top-story"&gt;#&lt;/a&gt;&lt;/h2&gt;
&lt;p&gt;&lt;strong&gt;&lt;a href="https://newsletter.semianalysis.com/p/finding-miscompiles-for-fun-not-profit"&gt;Finding Miscompiles for Fun, Not Profit&lt;/a&gt;&lt;/strong&gt;
A former Google and OpenAI compiler engineer threw $10,000 in API credits at Claude and ChatGPT to fuzz LLVM and NVIDIA&amp;rsquo;s &lt;code&gt;ptxas&lt;/code&gt;, discovering hundreds of deeply concerning miscompiles at an alarming rate. The real signal here isn&amp;rsquo;t just that AI can find bugs, but that &amp;ldquo;with enough subagents, all bugs are shallow&amp;rdquo;—a shift that makes elite-level code inspection simply a matter of having a massive compute budget.&lt;/p&gt;</description></item></channel></rss>