<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Agents on Hrushikesh Dokala</title><link>https://hrushikesh.dev/tags/agents/</link><description>Recent content in Agents on Hrushikesh Dokala</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Wed, 10 Jun 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://hrushikesh.dev/tags/agents/index.xml" rel="self" type="application/rss+xml"/><item><title>twin - turn your scattered agent sessions into a portable operating-profile</title><link>https://hrushikesh.dev/notes/twin/</link><pubDate>Wed, 10 Jun 2026 00:00:00 +0000</pubDate><guid>https://hrushikesh.dev/notes/twin/</guid><description>&lt;p>











&lt;figure class="">

 &lt;div class="img-container" >
 &lt;img loading="lazy" alt="twin" src="https://hrushikesh.dev/images/twin-hero.svg" >
 &lt;/div>

 
&lt;/figure>
&lt;/p>
&lt;p>ive been experimenting different approaches to bring &lt;strong>together&lt;/strong>, the &lt;strong>context across all my harnesses, which can be portable&lt;/strong>. the problem came from - i use hermes as my personal orchestrator (tried openclaw for a few months but took a shift - had to manually move the learnings to hermes), which uses claude code, codex as coding agents to do any task. mainly, 90% of my time spent at work is on claude code.&lt;/p></description></item><item><title>context switching is all you need</title><link>https://hrushikesh.dev/notes/context-switching/</link><pubDate>Sat, 30 May 2026 00:00:00 +0000</pubDate><guid>https://hrushikesh.dev/notes/context-switching/</guid><description>&lt;p>











&lt;figure class="">

 &lt;div class="img-container" >
 &lt;img loading="lazy" alt="cc" src="https://hrushikesh.dev/images/cc-screen.png" >
 &lt;/div>

 
&lt;/figure>
&lt;/p>
&lt;p>i have 4+ terminals open right now. each running claude code in bypass permissions, each on a different project. two are running experiments in independent worktrees, two i&amp;rsquo;m babysitting to fix prod issues.&lt;/p>
&lt;p>i&amp;rsquo;m not really doing any of it. i&amp;rsquo;m orchestrating.&lt;/p>
&lt;p>&lt;strong>the old advice&lt;/strong>&lt;/p>
&lt;p>there&amp;rsquo;s a &lt;a href="https://justinwelsh.me/essays/context-switching">popular take&lt;/a> that calls context switching &amp;ldquo;the silent killer of productivity.&amp;rdquo; batch your tasks, turn on do-not-disturb, single-task your way to clarity.&lt;/p>
&lt;p>for years this was right. but it was right for a specific reason that nobody states out loud: context switching was expensive because &lt;em>you&lt;/em> were the one holding the context. deep in a problem, someone pings you, and you lose 23 minutes rebuilding the mental model in your head.&lt;/p></description></item></channel></rss>