<?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>Clint Sebastian</title>
    <link>https://clintsebastian.github.io/</link>
    <description>Recent content on Clint Sebastian</description>
    <generator>Hugo</generator>
    <language>en-us</language>
    <lastBuildDate>Tue, 18 Aug 2026 20:00:00 +0000</lastBuildDate>
    <atom:link href="https://clintsebastian.github.io/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Ambient Intelligence</title>
      <link>https://clintsebastian.github.io/posts/ambient-intelligence/</link>
      <pubDate>Tue, 18 Aug 2026 20:00:00 +0000</pubDate>
      <guid>https://clintsebastian.github.io/posts/ambient-intelligence/</guid>
      <description>Cloud models are too slow for real-time human interaction. We examine tiny interaction models, sub-10ms on-device sensory loops, and asynchronous cloud planner architectures.</description>
    </item>
    <item>
      <title> Attention: Undivided &amp; Uncompressed</title>
      <link>https://clintsebastian.github.io/posts/undivided-attention/</link>
      <pubDate>Sun, 16 Aug 2026 14:30:00 +0000</pubDate>
      <guid>https://clintsebastian.github.io/posts/undivided-attention/</guid>
      <description>Nearly ten years after the &amp;#34;Attention Is All You Need&amp;#34; paper, attention still powers the most capable LLMs in the world. We take a look at how attention has evolved in the last decade, from its kernel geometry to the silicon it runs on. </description>
    </item>
  </channel>
</rss>
