<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Production-Performance on No Semicolons</title><link>https://nosemicolons.com/tags/production-performance/</link><description>Recent content in Production-Performance on No Semicolons</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Mon, 27 Apr 2026 10:08:39 +0000</lastBuildDate><atom:link href="https://nosemicolons.com/tags/production-performance/index.xml" rel="self" type="application/rss+xml"/><item><title>The AI Code Monitoring Playbook: How to Track Generated Code Performance in Production</title><link>https://nosemicolons.com/posts/ai-code-monitoring-production-performance-tracking/</link><pubDate>Mon, 27 Apr 2026 10:08:39 +0000</pubDate><guid>https://nosemicolons.com/posts/ai-code-monitoring-production-performance-tracking/</guid><description>&lt;p>Ever deployed AI-generated code only to wake up at 3 AM to a cascade of alerts? Yeah, me too. That perfect function Claude wrote for you might work beautifully in development, but production has a funny way of exposing edge cases that even the smartest AI models miss.&lt;/p>
&lt;p>The thing is, AI-generated code needs different monitoring than code we write ourselves. When I write a janky algorithm, I know exactly where the performance bottlenecks might lurk. But when GPT-4 generates an elegant solution I barely understand? That&amp;rsquo;s where things get interesting—and potentially problematic.&lt;/p></description></item></channel></rss>