<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Verification on oceanheart.ai</title>
    <link>https://www.oceanheart.ai/tags/verification/</link>
    <description>Recent content in Verification on oceanheart.ai</description>
    <generator>Hugo</generator>
    <language>en-gb</language>
    <lastBuildDate>Mon, 30 Mar 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://www.oceanheart.ai/tags/verification/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Sortie</title>
      <link>https://www.oceanheart.ai/projects/sortie/</link>
      <pubDate>Mon, 30 Mar 2026 00:00:00 +0000</pubDate>
      <guid>https://www.oceanheart.ai/projects/sortie/</guid>
      <description>&lt;h2 id=&#34;what-it-is&#34;&gt;What it is&lt;/h2&gt;&#xA;&lt;p&gt;An async adversarial code review system for multi-agent development workflows. Sortie runs a configurable roster of language models (Claude, Codex, Gemini) against a worker&amp;rsquo;s diff in parallel, synthesises their findings through a 4th-model debrief invocation, triages by severity, and gates worktree merges. All findings, dispositions, and cost data are captured in a structured ledger.&lt;/p&gt;&#xA;&lt;h2 id=&#34;why-it-exists&#34;&gt;Why it exists&lt;/h2&gt;&#xA;&lt;p&gt;Multi-agent swarm workflows produce code fast. They also produce unreviewed merges fast. The question isn&amp;rsquo;t &amp;ldquo;did the code pass tests&amp;rdquo; — it&amp;rsquo;s &amp;ldquo;did independent reviewers with different blind spots converge on the same issues?&amp;rdquo;&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
