<?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>Ecs on orndorff.dev</title>
    <link>https://orndorff.dev/tags/ecs/</link>
    <description>Recent content in Ecs on orndorff.dev</description>
    <generator>Hugo -- 0.138.0</generator>
    <language>en-us</language>
    <lastBuildDate>Thu, 09 Apr 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://orndorff.dev/tags/ecs/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Small LLMs, Big Reasoning: How a Neuro-Symbolic Expert System Makes Haiku Agents Reliable</title>
      <link>https://orndorff.dev/posts/neuro-symbolic-agents-nexus/</link>
      <pubDate>Thu, 09 Apr 2026 00:00:00 +0000</pubDate>
      <guid>https://orndorff.dev/posts/neuro-symbolic-agents-nexus/</guid>
      <description>&lt;p&gt;There&amp;rsquo;s a dirty secret in the AI agent space: most agent frameworks hand the model a bag of tools and pray. The model decides what to query, how to reason about results, and what conclusions to draw. For demos, this works great. For anything you&amp;rsquo;d actually bet your job on — compliance audits, student intervention decisions, infrastructure monitoring — it&amp;rsquo;s a liability.&lt;/p&gt;
&lt;p&gt;What if the model didn&amp;rsquo;t have to reason at all?&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
