<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Performance Regression on No Semicolons</title><link>https://nosemicolons.com/tags/performance-regression/</link><description>Recent content in Performance Regression on No Semicolons</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Mon, 25 May 2026 11:45:25 +0000</lastBuildDate><atom:link href="https://nosemicolons.com/tags/performance-regression/index.xml" rel="self" type="application/rss+xml"/><item><title>The AI Code Generation Performance Regression: How Generated Code Gets Slower Over Time (And 4 Optimization Strategies That Actually Work)</title><link>https://nosemicolons.com/posts/ai-code-generation-performance-regression-optimization-strategies/</link><pubDate>Mon, 25 May 2026 11:45:25 +0000</pubDate><guid>https://nosemicolons.com/posts/ai-code-generation-performance-regression-optimization-strategies/</guid><description>&lt;p>Ever noticed how that slick AI-generated function that blazed through your tests last month now crawls like it&amp;rsquo;s running through molasses? You&amp;rsquo;re not imagining things. There&amp;rsquo;s a sneaky phenomenon I&amp;rsquo;ve been tracking across dozens of projects: AI-generated code has a tendency to accumulate performance debt faster than human-written code.&lt;/p>
&lt;p>After digging into this pattern for the past year, I&amp;rsquo;ve discovered some fascinating insights about why this happens and, more importantly, what we can do about it. Let me share what I&amp;rsquo;ve learned about keeping AI-generated code fast and efficient.&lt;/p></description></item></channel></rss>