
  <rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
    <channel>
      <title>Karan Prasad - Founder, Obvix Labs</title>
      <link>https://karanprasad.com/blog</link>
      <description>Founder of Obvix Labs, an AI research lab building systems people can rely on. I write about what we&#39;re learning - AI safety, backend architecture, and what it takes to ship real software.</description>
      <language>en-us</language>
      <managingEditor>hello@karanprasad.com (Karan Prasad)</managingEditor>
      <webMaster>hello@karanprasad.com (Karan Prasad)</webMaster>
      <lastBuildDate>Mon, 06 Jul 2026 00:00:00 GMT</lastBuildDate>
      <atom:link href="https://karanprasad.com/tags/agents/feed.xml" rel="self" type="application/rss+xml"/>
      
  <item>
    <guid>https://karanprasad.com/blog/zapier-800-agents</guid>
    <title>I Investigated How Zapier Runs 800 AI Agents. Here&#39;s What the Headline Number Misses.</title>
    <link>https://karanprasad.com/blog/zapier-800-agents</link>
    <description>Everyone sees &#39;Zapier runs 800 agents&#39; and thinks the secret is multi-agent orchestration. After reading every source — the full podcast, the Stanford paper, the Spotify engineering blog, the HN failure threads — I found something different. The companies winning at agents spent years building platforms before AI arrived. The model was the last piece.</description>
    <pubDate>Mon, 06 Jul 2026 00:00:00 GMT</pubDate>
    <author>hello@karanprasad.com (Karan Prasad)</author>
    <category>ai</category><category>agents</category><category>platform-engineering</category><category>research</category>
  </item>

    </channel>
  </rss>
