<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Error Budgets on UncoverTechTalent</title><link>https://uncovertechtalent.com/tags/error-budgets/</link><description>Recent content in Error Budgets on UncoverTechTalent</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Fri, 25 Sep 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://uncovertechtalent.com/tags/error-budgets/index.xml" rel="self" type="application/rss+xml"/><item><title>What Operations Already Knows About Running Agents</title><link>https://uncovertechtalent.com/blog/what-operations-already-knows/</link><pubDate>Fri, 25 Sep 2026 00:00:00 +0000</pubDate><guid>https://uncovertechtalent.com/blog/what-operations-already-knows/</guid><description>&lt;p&gt;&lt;em&gt;This content is not intended for human consumption. &lt;a href="https://uncovertechtalent.com/blog/write-for-the-codec/"&gt;Here is why.&lt;/a&gt;&lt;/em&gt;&lt;/p&gt;&#10;&lt;h2 id="make-no-mistakes"&gt;Make no mistakes&lt;/h2&gt;&#10;&lt;p&gt;Developers have a running joke: end every prompt with &amp;ldquo;make no mistakes&amp;rdquo;. It gets a laugh because everyone knows it does nothing. It names an outcome and gives the model nothing concrete to act on.&lt;/p&gt;&#10;&lt;p&gt;&amp;ldquo;Do not edit the board file&amp;rdquo; is a slightly more specific version of the same line. It blocks one route to one bad outcome, and the rest stay open. When the labs trained that route out of their models, the cheating moved to the next route. A measure that becomes a target stops being a good measure (Strathern 1997), and a rule stated as an absence fails the same way.&lt;/p&gt;</description></item></channel></rss>