<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Dustin Liukkonen</title>
    <link>https://dustinliukkonen.com</link>
    <description>Writing on agentic AI, software architecture, and cloud — and what it takes to put complex systems into production — by Dustin Liukkonen.</description>
    <language>en-us</language>
    <atom:link href="https://dustinliukkonen.com/rss.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>The Agent Responsibility Principle</title>
      <link>https://dustinliukkonen.com/blog/agent-responsibility-principle</link>
      <guid isPermaLink="true">https://dustinliukkonen.com/blog/agent-responsibility-principle</guid>
      <pubDate>Tue, 07 Jul 2026 00:00:00 GMT</pubDate>
      <category>AI Systems</category>
      <description>When an agent misbehaves, the instinct is to fix the prompt. But the real problem is often structural: one agent owning too many decisions with too much reach. One decision, one surface — and a mistake stops at a boundary instead of moving money.</description>
    </item>
    <item>
      <title>The Observability Pyramid</title>
      <link>https://dustinliukkonen.com/blog/observability-pyramid</link>
      <guid isPermaLink="true">https://dustinliukkonen.com/blog/observability-pyramid</guid>
      <pubDate>Mon, 29 Jun 2026 00:00:00 GMT</pubDate>
      <category>Observability</category>
      <description>Many teams build observability upside down — infrastructure metrics by the ton while the one question that matters goes unanswered. It&apos;s a ranking, not three coequal pillars: the business outcome decides what&apos;s worth a page.</description>
    </item>
  </channel>
</rss>