<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: Jack</title>
    <description>The latest articles on DEV Community by Jack (@austriasoftwaroftwaredeveloper).</description>
    <link>https://dev.to/austriasoftwaroftwaredeveloper</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F96263%2F8dcd12df-b367-47d2-a0a4-cc749cb06ec8.png</url>
      <title>DEV Community: Jack</title>
      <link>https://dev.to/austriasoftwaroftwaredeveloper</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/austriasoftwaroftwaredeveloper"/>
    <language>en</language>
    <item>
      <title>Most "AI agent" demos work because nobody's actually using them.</title>
      <dc:creator>Jack</dc:creator>
      <pubDate>Sun, 03 May 2026 14:15:21 +0000</pubDate>
      <link>https://dev.to/austriasoftwaroftwaredeveloper/most-ai-agent-demos-work-because-nobodys-actually-using-them-5bc7</link>
      <guid>https://dev.to/austriasoftwaroftwaredeveloper/most-ai-agent-demos-work-because-nobodys-actually-using-them-5bc7</guid>
      <description>&lt;p&gt;Most "AI agent" demos work because exactly one person is using them — usually the person who built them.&lt;/p&gt;

&lt;p&gt;Production is different. Real users send malformed inputs, the API rate-limits, the model picks the wrong tool, the vector store returns stale results on day 90, and somebody asks for a feature your prompt scaffold can't bend around.&lt;/p&gt;

&lt;p&gt;Half my client work right now is turning agent prototypes into things that survive contact with actual users. The unsexy parts — retries, idempotency, eval suites, observability, structured tool I/O — are 80% of the real build.&lt;/p&gt;

&lt;p&gt;If your agent works in the demo and breaks in prod, the demo wasn't the product. The retries were.&lt;/p&gt;

</description>
      <category>ai</category>
    </item>
    <item>
      <title>Most "AI agent" demos work because exactly one person is using them — usually the person who built them.

Production is different. Real users send malformed inputs, the API rate-limits, the model picks the wrong tool, the vector store returns stale results</title>
      <dc:creator>Jack</dc:creator>
      <pubDate>Sun, 03 May 2026 14:13:51 +0000</pubDate>
      <link>https://dev.to/austriasoftwaroftwaredeveloper/most-ai-agent-demos-work-because-exactly-one-person-is-using-them-usually-the-person-who-built-4nb</link>
      <guid>https://dev.to/austriasoftwaroftwaredeveloper/most-ai-agent-demos-work-because-exactly-one-person-is-using-them-usually-the-person-who-built-4nb</guid>
      <description></description>
      <category>agents</category>
      <category>ai</category>
      <category>llm</category>
      <category>softwareengineering</category>
    </item>
  </channel>
</rss>
