<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Voice on dolftax</title>
    <link>https://dolftax.com/tags/voice/</link>
    <description>Recent content in Voice on dolftax</description>
    <generator>Hugo</generator>
    <language>en-us</language>
    <lastBuildDate>Mon, 06 Apr 2026 18:35:00 +0000</lastBuildDate>
    <atom:link href="https://dolftax.com/tags/voice/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Agents work best with more context, voice enables that</title>
      <link>https://dolftax.com/notes/1775525700/</link>
      <pubDate>Mon, 06 Apr 2026 18:35:00 +0000</pubDate>
      <guid>https://dolftax.com/notes/1775525700/</guid>
      <description>&lt;p&gt;&amp;ldquo;Fix the checkout bug&amp;rdquo;. That&amp;rsquo;s the kind of prompt agents usually get. Agents can pull logs, check deployments, read support tickets &amp;ndash; that part is solved. What they can&amp;rsquo;t pull is what you actually want. The tradeoffs you care about, the way you want it done.&lt;/p&gt;&#xA;&lt;p&gt;When you talk, you don&amp;rsquo;t filter as much. You ramble. You over-explain. You mention things you weren&amp;rsquo;t sure were relevant. Voice gets people to stop editing themselves before hitting enter.&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
