<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Junming Cao — Blogs</title><description>Articles by Junming Cao.</description><link>https://jamescao2048.github.io/</link><language>en</language><item><title>When Coding Gets Cheaper, Choosing What to Build Matters More</title><link>https://jamescao2048.github.io/writing/choosing-work-with-coding-agents/</link><guid isPermaLink="true">https://jamescao2048.github.io/writing/choosing-work-with-coding-agents/</guid><description>How milestones, architecture decision records, and task specifications helped me keep fast-moving coding agents focused on the next release.</description><pubDate>Mon, 15 Jun 2026 03:54:00 GMT</pubDate></item><item><title>Prompt Rules Need Runtime Enforcement</title><link>https://jamescao2048.github.io/writing/prompt-rules-runtime-enforcement/</link><guid isPermaLink="true">https://jamescao2048.github.io/writing/prompt-rules-runtime-enforcement/</guid><description>A compute-kernel example of moving state, retries, and acceptance checks from prompt instructions into the software around a coding agent.</description><pubDate>Mon, 25 May 2026 03:57:00 GMT</pubDate></item><item><title>Why Autonomous Research Loops Do Not Automatically Build Useful Apps</title><link>https://jamescao2048.github.io/writing/autonomous-research-feedback-loops/</link><guid isPermaLink="true">https://jamescao2048.github.io/writing/autonomous-research-feedback-loops/</guid><description>Research agents benefit from stable goals and fast, automatic feedback. Application development needs the same foundations, plus an explicit boundary around what tests cannot decide.</description><pubDate>Sat, 02 May 2026 03:55:00 GMT</pubDate></item></channel></rss>