<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Problem on preconfiguration.com</title>
    <link>https://preconfiguration.com/tags/problem/</link>
    <description>Recent content in Problem on preconfiguration.com</description>
    <generator>Hugo</generator>
    <language>en-us</language>
    <lastBuildDate>Tue, 29 Sep 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://preconfiguration.com/tags/problem/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>The Problem Preconfiguration Solves: Coding Agents Start Every Task on a Machine Nobody Set Up</title>
      <link>https://preconfiguration.com/the-problem-preconfiguration-solves-coding-agents-start-every-task-on-a-machine-nobody-set-up/</link>
      <pubDate>Tue, 29 Sep 2026 00:00:00 +0000</pubDate>
      <guid>https://preconfiguration.com/the-problem-preconfiguration-solves-coding-agents-start-every-task-on-a-machine-nobody-set-up/</guid>
      <description>&lt;p&gt;AI coding agents start each task on a blank machine. When that machine isn&amp;rsquo;t set up, they guess, fail or hand back untested code, and every vendor wants the setup written in its own file.&lt;/p&gt;&#xA;&lt;h2 id=&#34;agents-are-poor-at-doing-it-alone&#34;&gt;Agents Are Poor at Doing It Alone&lt;/h2&gt;&#xA;&lt;p&gt;If agents could reliably set up a repository by themselves, the problem would solve itself. The benchmarks say they can&amp;rsquo;t, yet:&lt;/p&gt;&#xA;&lt;p&gt;&lt;img src=&#34;https://preconfiguration.com/images/the-problem.png&#34; alt=&#34;How often the best approach in each benchmark set up a repository on its own: 6.69% of 329 Python repositories and 29.47% of 665 JVM repositories in EnvBench, from JetBrains Research (2025), and a basic CPU run in 17 of the 29 research repositories that have one (58.6%) in ResearchEnvBench (2026).&#34;&gt;&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
