<?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>Deployment on AI Notes</title>
    <link>https://ai.ngchi.online/tags/deployment/</link>
    <description>Recent content in Deployment on AI Notes</description>
    <generator>Hugo</generator>
    <language>en-us</language>
    <lastBuildDate>Wed, 30 Sep 2026 13:00:51 +0700</lastBuildDate>
    <atom:link href="https://ai.ngchi.online/tags/deployment/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Stripe Says Kai Runs on 1,000 Skills. Its Partner Says Quality Drops Past 150</title>
      <link>https://ai.ngchi.online/posts/stripe-says-kai-runs-on-1000-skills-its-partner-says-quality-drops-past-150/</link>
      <pubDate>Wed, 30 Sep 2026 13:00:51 +0700</pubDate>
      <guid>https://ai.ngchi.online/posts/stripe-says-kai-runs-on-1000-skills-its-partner-says-quality-drops-past-150/</guid>
      <description>Stripe says its internal agent is connected to 1,000+ skills. Its infrastructure vendor reports quality degrading past 150. A close read of both posts.</description>
    </item>
    <item>
      <title>Muse Is Built as If the Agent Were Already Compromised</title>
      <link>https://ai.ngchi.online/posts/muse-is-built-as-if-the-agent-were-already-compromised/</link>
      <pubDate>Mon, 28 Sep 2026 17:31:00 +0700</pubDate>
      <guid>https://ai.ngchi.online/posts/muse-is-built-as-if-the-agent-were-already-compromised/</guid>
      <description>Meta&amp;rsquo;s engineering write-up for Muse states its premise in one line: two isolated security domains on one box, not an LLM powered agent with root. A walk through what that means in practice.</description>
    </item>
  </channel>
</rss>
