<?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: Cogweald</title>
    <description>The latest articles on DEV Community by Cogweald (@jones_chen_592558a62a5c44).</description>
    <link>https://dev.to/jones_chen_592558a62a5c44</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.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F4121648%2F49154cbf-d219-429e-8281-d95b25b4cee6.png</url>
      <title>DEV Community: Cogweald</title>
      <link>https://dev.to/jones_chen_592558a62a5c44</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/jones_chen_592558a62a5c44"/>
    <language>en</language>
    <item>
      <title>Most AI products forget. Cogweald is designed to remember.</title>
      <dc:creator>Cogweald</dc:creator>
      <pubDate>Sat, 12 Sep 2026 04:23:58 +0000</pubDate>
      <link>https://dev.to/jones_chen_592558a62a5c44/most-ai-products-forget-cogweald-is-designed-to-remember-383d</link>
      <guid>https://dev.to/jones_chen_592558a62a5c44/most-ai-products-forget-cogweald-is-designed-to-remember-383d</guid>
      <description>&lt;p&gt;Many AI demos feel intelligent until you ask the same question the next day.&lt;/p&gt;

&lt;p&gt;The problem is often not the model. The product simply has no durable world state. The conversation ends, the memory ends with it, and the next request starts from a blank page.&lt;/p&gt;

&lt;p&gt;Cogweald takes a different approach: put AI inside a world that keeps running. Characters have state. Events leave consequences. Unresolved threads remain active. What happened yesterday constrains what can happen today.&lt;/p&gt;

&lt;p&gt;This is more than adding a &lt;code&gt;memory&lt;/code&gt; table. The hard questions are:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Which information deserves to survive?&lt;/li&gt;
&lt;li&gt;Which information is no longer valid?&lt;/li&gt;
&lt;li&gt;How do we show the model the right context before its next decision?&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;We are not trying to build another chat window. We are exploring what an AI system looks like when it can accumulate history.&lt;/p&gt;

&lt;p&gt;What should your AI product never forget?&lt;/p&gt;

</description>
      <category>ai</category>
      <category>agents</category>
      <category>architecture</category>
      <category>buildinpublic</category>
    </item>
  </channel>
</rss>
