<?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>PromptPotter — articles</title>
    <link>https://promptpotter.com</link>
    <description>The article series behind PromptPotter: program evolution with an LLM as the proposer, what it took to make it a tool, and what attaches to the loop next.</description>
    <language>en</language>
    <atom:link href="https://promptpotter.com/feed.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>The loop is the floor.</title>
      <link>https://promptpotter.com/headroom</link>
      <guid isPermaLink="true">https://promptpotter.com/headroom</guid>
      <pubDate>Wed, 12 Aug 2026 00:00:00 GMT</pubDate>
      <dc:creator>David Streuli</dc:creator>
      <description>The optimizing loop is the floor, not the ceiling. Everything worth adding attaches to it rather than replacing it: self-improvement, synthetic data, topology mapping, bigger networks, web search. How much any of them adds is not yet measured.</description>
    </item>
    <item>
      <title>The science was done. The tool was missing.</title>
      <link>https://promptpotter.com/zero-friction</link>
      <guid isPermaLink="true">https://promptpotter.com/zero-friction</guid>
      <pubDate>Sun, 09 Aug 2026 00:00:00 GMT</pubDate>
      <dc:creator>David Streuli</dc:creator>
      <description>Why prompt optimization stayed in the lab: not the loop or the statistics, but the hours between reading about a result and having one running on your own work. What it takes to close that gap.</description>
    </item>
    <item>
      <title>Simple but Effective Search: Entering the Era Where It Climbs with the Models.</title>
      <link>https://promptpotter.com/program-evolution</link>
      <guid isPermaLink="true">https://promptpotter.com/program-evolution</guid>
      <pubDate>Sat, 13 Jun 2026 00:00:00 GMT</pubDate>
      <dc:creator>David Streuli</dc:creator>
      <description>Evolutionary search is an old, universal loop: biology, genetic algorithms, RL, AutoML. AlphaEvolve and PromptPotter are that same loop with an LLM as the proposer, which finally makes it general enough to improve almost any artifact.</description>
    </item>
  </channel>
</rss>
