<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
<channel>
  <title>Awesomist blog</title>
  <link>https://awesomist.io/blog</link>
  <atom:link href="https://awesomist.io/blog/feed.xml" rel="self" type="application/rss+xml"/>
  <description>Practical writing on AI strategy, agentic AI and software engineering.</description>
  <language>en</language>
  <item>
    <title>RAG or fine-tuning: how to choose for a business use case</title>
    <link>https://awesomist.io/blog/rag-or-fine-tuning</link>
    <guid>https://awesomist.io/blog/rag-or-fine-tuning</guid>
    <pubDate>Wed, 14 Oct 2026 00:00:00 GMT</pubDate>
    <description>Retrieval and fine-tuning solve different problems. A plain guide to when each one fits, when to combine them, and how to decide with evidence.</description>
  </item>
  <item>
    <title>Why AI pilots stall before production, and how to get one live</title>
    <link>https://awesomist.io/blog/why-ai-pilots-stall</link>
    <guid>https://awesomist.io/blog/why-ai-pilots-stall</guid>
    <pubDate>Wed, 14 Oct 2026 00:00:00 GMT</pubDate>
    <description>Most AI pilots never reach daily use. The causes are usually the same few, and almost none of them are about the model. Here is what to plan for from the start.</description>
  </item>
</channel>
</rss>
