
  <rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
    <channel>
      <title>Jacopo Daeli</title>
      <link>https://jacopodaeli.com/blog</link>
      <description>Distinguished Engineer at Palo Alto Networks. Writing about machine learning, distributed systems, and cybersecurity.</description>
      <language>en-US</language>
      <managingEditor>hello@jacopodaeli.com (Jacopo Daeli)</managingEditor>
      <webMaster>hello@jacopodaeli.com (Jacopo Daeli)</webMaster>
      <lastBuildDate>Mon, 11 May 2026 00:00:00 GMT</lastBuildDate>
      <atom:link href="https://jacopodaeli.com/feed.xml" rel="self" type="application/rss+xml"/>
      
  <item>
    <guid>https://jacopodaeli.com/blog/ai-platform-wrote-its-own-benchmark</guid>
    <title>The AI Platform That Wrote Its Own Benchmark Won</title>
    <link>https://jacopodaeli.com/blog/ai-platform-wrote-its-own-benchmark</link>
    <description>A new 68-metric benchmark tests whether AI app-building platforms can function as complete software agencies. The findings on frontend-backend decoupling and universal security failures are real, but two of three authors work for the highest-scoring platform, and the conflict deserves scrutiny.</description>
    <pubDate>Mon, 11 May 2026 00:00:00 GMT</pubDate>
    <author>hello@jacopodaeli.com (Jacopo Daeli)</author>
  </item>

    </channel>
  </rss>
