<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Software on The Coders Blog</title><link>https://thecodersblog.com/tag/software/</link><description>Recent content in Software on The Coders Blog</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Thu, 07 May 2026 07:44:29 +0000</lastBuildDate><atom:link href="https://thecodersblog.com/tag/software/index.xml" rel="self" type="application/rss+xml"/><item><title>ProgramBench: Can AI Rebuild Software?</title><link>https://thecodersblog.com/language-models-rebuilding-programs-2026/</link><pubDate>Thu, 07 May 2026 07:44:29 +0000</pubDate><guid>https://thecodersblog.com/language-models-rebuilding-programs-2026/</guid><description>&lt;p&gt;Imagine handing over a compiled program, its documentation, and saying, &amp;ldquo;Rebuild this.&amp;rdquo; Not by looking at the source, not by searching the web, but by understanding the &lt;em&gt;essence&lt;/em&gt; of what it does and recreating it from scratch. This isn&amp;rsquo;t a hypothetical for the future; it&amp;rsquo;s the challenge posed by ProgramBench, a new benchmark designed to stress-test the current frontier of AI agents and language models in software creation. The results? Frankly, they’re a stark reminder of how far we still have to go.&lt;/p&gt;</description></item></channel></rss>