<?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>Automation on Po-Wei Chen, MD — Physiatrist &amp; Builder</title><link>https://drpwchen.com/en/tags/automation/</link><description>Recent content in Automation on Po-Wei Chen, MD — Physiatrist &amp; Builder</description><image><title>Po-Wei Chen, MD — Physiatrist &amp; Builder</title><url>https://drpwchen.com/og-default.png</url><link>https://drpwchen.com/og-default.png</link></image><generator>Hugo</generator><language>en</language><lastBuildDate>Sat, 25 Jul 2026 06:45:00 +0800</lastBuildDate><atom:link href="https://drpwchen.com/en/tags/automation/index.xml" rel="self" type="application/rss+xml"/><item><title>AI running out of gas halfway through a midnight task? I built it a fuel gauge and automatic brakes</title><link>https://drpwchen.com/en/posts/claude-pacer/</link><pubDate>Sat, 25 Jul 2026 06:45:00 +0800</pubDate><guid>https://drpwchen.com/en/posts/claude-pacer/</guid><description>The most common way for AI to die running tasks autonomously at night isn&amp;#39;t bad code, it&amp;#39;s hitting the usage limit and getting cut off. claude-pacer solves this with a three-layer design: a status line for humans, a budget alert for AI, and a GO/PACE/STOP verdict for agents to run. Open source MIT.</description></item><item><title>My Open-Source Tool Was Targeted by a Bot Marketer!</title><link>https://drpwchen.com/en/posts/github-bot-marketing/</link><pubDate>Thu, 09 Jul 2026 10:45:00 +0800</pubDate><guid>https://drpwchen.com/en/posts/github-bot-marketing/</guid><description>My open-source paper tool received a GitHub issue full of sincere praise. After checking, I found it was automated marketing spammed by a bot to 26 repos within two minutes. This post breaks down its tactics, the paid marketplace model behind it, and why running it with your own subscription is a better deal.</description></item><item><title>I Open-Sourced My Paper Reading Radar</title><link>https://drpwchen.com/en/posts/paper-radar/</link><pubDate>Wed, 01 Jul 2026 06:45:00 +0800</pubDate><guid>https://drpwchen.com/en/posts/paper-radar/</guid><description>I took my clunky paper RSS setup and turned it into an online site that automatically fetches papers daily, ranks them by my interest, and learns my tastes over time. Selected papers are then automatically fetched, digested, and critically appraised. Now I&amp;#39;ve organized it into an open-source version on GitHub. Not just for medicine—any field with RSS feeds can fork it to build their own radar.</description></item><item><title>I installed a money-saving dashboard for my AI</title><link>https://drpwchen.com/en/posts/ai-token-4-budget-dashboard/</link><pubDate>Mon, 29 Jun 2026 06:30:00 +0800</pubDate><guid>https://drpwchen.com/en/posts/ai-token-4-budget-dashboard/</guid><description>The final part of the saving tokens series: not everything needs the smartest brain, and even for reading images, the cheapest model is often enough. Toss noisy chores to a clone to keep your desk clean, then install a dashboard, or even let the AI watch its own usage and pace the budget just right.</description></item><item><title>If an abacus works, don't boot up the supercomputer: When to actually make AI think?</title><link>https://drpwchen.com/en/posts/ai-token-2-script-vs-llm/</link><pubDate>Mon, 29 Jun 2026 06:30:00 +0800</pubDate><guid>https://drpwchen.com/en/posts/ai-token-2-script-vs-llm/</guid><description>Some jobs simply shouldn&amp;#39;t be given to AI. Clear-cut, repetitive tasks belong to hardcoded scripts—they&amp;#39;re fast, accurate, and practically free. Vague, judgment-heavy tasks are what you boot up AI for. Here&amp;#39;s how I read my medical textbook PDFs as an example of making this tradeoff.</description></item><item><title>Organizing a clean desk for the AI</title><link>https://drpwchen.com/en/posts/ai-token-3-clean-desk/</link><pubDate>Mon, 29 Jun 2026 06:30:00 +0800</pubDate><guid>https://drpwchen.com/en/posts/ai-token-3-clean-desk/</guid><description>There&amp;#39;s a cost you pay with every sentence: the config file residing right in front of the AI. Keep it lean, layer the memory, make good use of prompt caching to bill repeated content at one-tenth the price, and throw away the noise of successful commands directly. A clean desk doesn&amp;#39;t just save money; the AI answers more accurately, too.</description></item><item><title>Built a PM&amp;R Continuing Education Radar</title><link>https://drpwchen.com/en/posts/rehab-course-radar/</link><pubDate>Wed, 24 Jun 2026 07:30:00 +0800</pubDate><guid>https://drpwchen.com/en/posts/rehab-course-radar/</guid><description>I built a website, courses.drpwchen.com, that automatically aggregates courses and announcements from 17 rehabilitation-related societies and institutions. It updates every 3 hours, supports RSS or ntfy notifications, and is free with no login required.</description></item></channel></rss>