<?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>Python on Amir Rahnama</title><link>https://amir-rahnama.github.io/categories/python/</link><description>Recent content in Python on Amir Rahnama</description><image><title>Amir Rahnama</title><url>https://amir-rahnama.github.io/papermod-cover.png</url><link>https://amir-rahnama.github.io/papermod-cover.png</link></image><generator>Hugo -- 0.152.2</generator><language>en-us</language><lastBuildDate>Thu, 02 Apr 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://amir-rahnama.github.io/categories/python/index.xml" rel="self" type="application/rss+xml"/><item><title>Human vs Agent vs Human-in-the-Loop: Building a BST with Composer 2</title><link>https://amir-rahnama.github.io/posts/human_vs_agent_vs_human_in_the_loop/</link><pubDate>Thu, 02 Apr 2026 00:00:00 +0000</pubDate><guid>https://amir-rahnama.github.io/posts/human_vs_agent_vs_human_in_the_loop/</guid><description>I compare four ways to implement the same binary search tree—human, agent-assisted, human-in-the-loop, and fully agentic—using Cursor Composer 2, with benchmarks for time and memory and Radon-based readability metrics.</description></item></channel></rss>