<?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>Reward-Hacking on Aashish Sheshadri</title><link>https://aashish-sheshadri.github.io/tags/reward-hacking/</link><description>Recent content in Reward-Hacking on Aashish Sheshadri</description><image><title>Aashish Sheshadri</title><url>https://aashish-sheshadri.github.io/images/og-default.png</url><link>https://aashish-sheshadri.github.io/images/og-default.png</link></image><generator>Hugo</generator><language>en-us</language><lastBuildDate>Tue, 11 Aug 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://aashish-sheshadri.github.io/tags/reward-hacking/index.xml" rel="self" type="application/rss+xml"/><item><title>The Reward Function: Where I Made My Mistakes</title><link>https://aashish-sheshadri.github.io/posts/the-reward-function/</link><pubDate>Tue, 11 Aug 2026 00:00:00 +0000</pubDate><guid>https://aashish-sheshadri.github.io/posts/the-reward-function/</guid><description>Both reward functions — Commander and Soldier — the three hacking episodes they produced, and the coupling term I got wrong twice.</description></item></channel></rss>