<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Structured-Output on Jamal Yusuf</title><link>https://jamal.dev/tags/structured-output/</link><description>Recent content in Structured-Output on Jamal Yusuf</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Sun, 14 Jun 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://jamal.dev/tags/structured-output/index.xml" rel="self" type="application/rss+xml"/><item><title>Validating Agent Outputs in Go Before They Touch Production</title><link>https://jamal.dev/writing/go-validate-agent-outputs/</link><pubDate>Sun, 14 Jun 2026 00:00:00 +0000</pubDate><guid>https://jamal.dev/writing/go-validate-agent-outputs/</guid><description>&lt;p&gt;An agent that sounds right is the most dangerous kind of wrong.&lt;/p&gt;
&lt;p&gt;I have seen it in healthcare workflows — confident JSON, clean prose, a downstream system that accepts the payload and only screams twenty minutes later when reconciliation fails. The model did its job. The &lt;strong&gt;system&lt;/strong&gt; failed to treat probability like probability.&lt;/p&gt;
&lt;p&gt;In Go, I fix that with a boring layer between the model and the world: &lt;strong&gt;validate before side effects&lt;/strong&gt;. Always.&lt;/p&gt;</description></item></channel></rss>