
  <rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
    <channel>
      <title>Rambling Me</title>
      <link>https://www.rambling-me.com/blog</link>
      <description>A blog to reflect my thoughts, learnings, and observations</description>
      <language>en-us</language>
      <managingEditor>me@rambling-me.com (Dopeking)</managingEditor>
      <webMaster>me@rambling-me.com (Dopeking)</webMaster>
      <lastBuildDate>Thu, 03 Sep 2026 00:00:00 GMT</lastBuildDate>
      <atom:link href="https://www.rambling-me.com/tags/evaluation/feed.xml" rel="self" type="application/rss+xml"/>
      
  <item>
    <guid>https://www.rambling-me.com/blog/posts/technology/agentic/decision-layer/3-the-eval-harness-you-already-have</guid>
    <title>Agentic Systems: The Eval Harness You Already Have</title>
    <link>https://www.rambling-me.com/blog/posts/technology/agentic/decision-layer/3-the-eval-harness-you-already-have</link>
    <description>Tests check whether code does what it was written to do. They have nothing to say about whether a judgement was any good. Scoring a non-deterministic system means comparing recorded predictions against observed outcomes, and most systems already emit both without recognising that they have the raw material for an eval harness. Why calibration matters more than accuracy, what bucketing is for, and the two problems - outcome lag and correlated error - that make this harder than it looks.</description>
    <pubDate>Thu, 03 Sep 2026 00:00:00 GMT</pubDate>
    <author>me@rambling-me.com (Dopeking)</author>
    <category>technology</category><category>agentic</category><category>multi-agent-systems</category><category>evaluation</category><category>reliability</category><category>observability</category><category>practitioner</category>
  </item>

    </channel>
  </rss>
