<?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>Evaluation on Dmytro Kovalchuk</title>
    <link>https://kdmytro.com/tags/evaluation/</link>
    <description>Recent content in Evaluation on Dmytro Kovalchuk</description>
    <generator>Hugo -- 0.147.0</generator>
    <language>en</language>
    <lastBuildDate>Tue, 10 Feb 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://kdmytro.com/tags/evaluation/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>LLM-as-Judge is Evolving. Meet Agent-as-Judge.</title>
      <link>https://kdmytro.com/posts/agent-as-judge/</link>
      <pubDate>Tue, 10 Feb 2026 00:00:00 +0000</pubDate>
      <guid>https://kdmytro.com/posts/agent-as-judge/</guid>
      <description>Why evaluating agents requires agents — from passive scoring to active investigation</description>
    </item>
    <item>
      <title>How Do You Actually Evaluate an AI Research Agent?</title>
      <link>https://kdmytro.com/posts/evaluating-ai-research-agents/</link>
      <pubDate>Tue, 14 Jan 2025 00:00:00 +0000</pubDate>
      <guid>https://kdmytro.com/posts/evaluating-ai-research-agents/</guid>
      <description>Beyond &amp;#39;it runs&amp;#39; — figuring out how to know if your AI agent is actually good</description>
    </item>
  </channel>
</rss>
