<?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>Evals on Miguel Lameiro | Cybersecurity Blog &amp; Security Writeups</title><link>https://blog.lameiro0x.com/tags/evals/</link><description>Recent content in Evals on Miguel Lameiro | Cybersecurity Blog &amp; Security Writeups</description><generator>Hugo -- 0.161.1</generator><language>en-us</language><lastBuildDate>Tue, 05 May 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://blog.lameiro0x.com/tags/evals/index.xml" rel="self" type="application/rss+xml"/><item><title>Automated Evals for LLMOps: Testing LLM Apps in CI</title><link>https://blog.lameiro0x.com/notes/ai-security/automated-evals-for-llmops-testing-llm-apps-in-ci/</link><pubDate>Tue, 05 May 2026 00:00:00 +0000</pubDate><guid>https://blog.lameiro0x.com/notes/ai-security/automated-evals-for-llmops-testing-llm-apps-in-ci/</guid><description>&lt;h2 id="introduction"&gt;Introduction&lt;/h2&gt;
&lt;p&gt;Traditional software tests usually compare a known input with a predictable output. LLM applications are different because the output is generated, variable, and sometimes correct in more than one form.&lt;/p&gt;
&lt;p&gt;That does not mean LLM apps cannot be tested. It means the test suite needs several layers: deterministic checks where possible, model-graded evaluations where judgment is required, hallucination checks against known context, and CI automation so regressions are caught before release.&lt;/p&gt;</description></item></channel></rss>