<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/">
  <channel>
    <title>AI on Barajas</title>
    <link>https://barajas.blog/tags/ai/</link>
    <description>Recent content in AI on Barajas</description>
    <generator>Hugo</generator>
    <language>en-US</language>
    <lastBuildDate>Wed, 01 Jul 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://barajas.blog/tags/ai/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>No, Qwen 27B Isn&#39;t Enough</title>
      <link>https://barajas.blog/posts/no-qwen-27b-isnt-enough/</link>
      <pubDate>Wed, 01 Jul 2026 00:00:00 +0000</pubDate>
      <guid>https://barajas.blog/posts/no-qwen-27b-isnt-enough/</guid>
      <description>A Hacker News thread said a 27 billion parameter model is all you need for real coding work. I&amp;rsquo;ve run that model against real codebases. It isn&amp;rsquo;t, and here&amp;rsquo;s the math on why.</description>
    </item>
  </channel>
</rss>
