<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Codex on Daffa Abhipraya</title><link>https://blog.abhipraya.dev/tags/codex/</link><description>Recent content in Codex on Daffa Abhipraya</description><generator>Hugo</generator><language>en-us</language><copyright>© Daffa Abhipraya</copyright><lastBuildDate>Fri, 13 Mar 2026 00:00:00 +0700</lastBuildDate><atom:link href="https://blog.abhipraya.dev/tags/codex/index.xml" rel="self" type="application/rss+xml"/><item><title>Three Coding Agents, One Brain</title><link>https://blog.abhipraya.dev/posts/agents-dev-setup-on-macbook/</link><pubDate>Fri, 13 Mar 2026 00:00:00 +0700</pubDate><guid>https://blog.abhipraya.dev/posts/agents-dev-setup-on-macbook/</guid><description>&lt;p>People keep asking how my AI dev setup actually works. Friends, classmates, engineers I bump into - most of them don&amp;rsquo;t really want a model leaderboard. They want the practical version: which tools I use, how they fit together, what runs locally on my MacBook, and how I keep the whole thing from drifting into chaos.&lt;/p>
&lt;p>Short answer: I don&amp;rsquo;t lean on one AI coding tool for everything. I run a shared workflow, shared instructions, shared skills, and shared commands across multiple clients. The three I reach for on the MacBook are &lt;a href="https://www.anthropic.com/product/claude-code">Claude Code&lt;/a>, &lt;a href="https://openai.com/codex/">Codex&lt;/a>, and &lt;a href="https://github.com/withpi/codemode">pi&lt;/a>. Around them sit the supporting pieces: specialist skills, MCP tools, voice dictation, dotfiles sync, and a workspace layer that keeps sessions isolated without making the environment feel fragmented.&lt;/p></description></item></channel></rss>