<?xml version="1.0" encoding="UTF-8"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Miao YU · Lab — Research Insights</title><description>Methodology, domain notes and practice retrospectives. Chinese is the original; English is an unreviewed AI draft.</description><link>https://lab.miao-yu.com/en/</link><language>en</language><atom:link href="https://lab.miao-yu.com/en/rss.xml" rel="self" type="application/rss+xml"/><item><title>murRipple, closed out — a song woven into visible ripples</title><link>https://lab.miao-yu.com/en/insights/murripple-closeout/</link><guid isPermaLink="true">https://lab.miao-yu.com/en/insights/murripple-closeout/</guid><description>The project is done. It leaves two documents behind, one for readers who want to understand it and one for people who want to use it.</description><pubDate>Thu, 20 Aug 2026 00:00:00 GMT</pubDate><category>Methodology</category></item><item><title>I had an AI run a capability checkup on me</title><link>https://lab.miao-yu.com/reports/ai-engineering-checkup-en.html</link><guid isPermaLink="true">https://lab.miao-yu.com/reports/ai-engineering-checkup-en.html</guid><description>One rule only. Nothing I claimed about myself could count, only machine-checkable traces in the repositories.</description><pubDate>Sat, 15 Aug 2026 00:00:00 GMT</pubDate><category>Methodology</category></item><item><title>Two Weeks from Zero to a Customer Demo</title><link>https://lab.miao-yu.com/en/insights/two-weeks-to-a-demo/</link><guid isPermaLink="true">https://lab.miao-yu.com/en/insights/two-weeks-to-a-demo/</guid><description>Speed does not come from working late — it comes from cutting the work down to a single line.</description><pubDate>Mon, 10 Aug 2026 00:00:00 GMT</pubDate><category>Methodology</category></item><item><title>Verified = Command + Output</title><link>https://lab.miao-yu.com/en/insights/verified-means-command-and-output/</link><guid isPermaLink="true">https://lab.miao-yu.com/en/insights/verified-means-command-and-output/</guid><description>The bill for one “it should be fine” usually falls due three weeks later.</description><pubDate>Mon, 10 Aug 2026 00:00:00 GMT</pubDate><category>Methodology</category></item><item><title>One Person Walks into Wind-Tower Monitoring</title><link>https://lab.miao-yu.com/en/insights/one-phone-number/</link><guid isPermaLink="true">https://lab.miao-yu.com/en/insights/one-phone-number/</guid><description>The hardest step is not the technology — it is getting someone to lend you their asset for an experiment.</description><pubDate>Mon, 10 Aug 2026 00:00:00 GMT</pubDate><category>Domain notes</category></item><item><title>Putting Reins on the AI</title><link>https://lab.miao-yu.com/en/insights/reins-not-conversation/</link><guid isPermaLink="true">https://lab.miao-yu.com/en/insights/reins-not-conversation/</guid><description>What sets the ceiling is not how you ask — it is the layer outside the model.</description><pubDate>Mon, 10 Aug 2026 00:00:00 GMT</pubDate><category>Methodology</category></item><item><title>Human + AI: A Multi-Window Operating System</title><link>https://lab.miao-yu.com/en/insights/multi-window-collaboration/</link><guid isPermaLink="true">https://lab.miao-yu.com/en/insights/multi-window-collaboration/</guid><description>Not hunting for a smarter model — designing an organization that keeps ordinary AIs from drifting.</description><pubDate>Sun, 09 Aug 2026 00:00:00 GMT</pubDate><category>Methodology</category></item><item><title>The City Under Your Feet Is Talking: An Introduction to DAS</title><link>https://lab.miao-yu.com/en/insights/das-explained/</link><guid isPermaLink="true">https://lab.miao-yu.com/en/insights/das-explained/</guid><description>How an ordinary telecom fiber becomes a stethoscope several kilometers long.</description><pubDate>Sun, 09 Aug 2026 00:00:00 GMT</pubDate><category>Domain notes</category></item><item><title>Signal and Noise</title><link>https://lab.miao-yu.com/en/insights/signal-and-noise/</link><guid isPermaLink="true">https://lab.miao-yu.com/en/insights/signal-and-noise/</guid><description>The discipline for deciding whether a waveform is a real event turns out to work on AI answers too.</description><pubDate>Sun, 09 Aug 2026 00:00:00 GMT</pubDate><category>Domain notes</category></item><item><title>From Optimal Transport to LLMs</title><link>https://lab.miao-yu.com/en/insights/optimal-transport-to-llm/</link><guid isPermaLink="true">https://lab.miao-yu.com/en/insights/optimal-transport-to-llm/</guid><description>After changing tracks three times, I found that what followed me was never the tools — it was one question.</description><pubDate>Sun, 09 Aug 2026 00:00:00 GMT</pubDate><category>Domain notes</category></item><item><title>Why I Write a “Task Book” for Every Project</title><link>https://lab.miao-yu.com/en/insights/task-books/</link><guid isPermaLink="true">https://lab.miao-yu.com/en/insights/task-books/</guid><description>Once AI made writing code easy, what matters most is saying exactly what you want.</description><pubDate>Sun, 09 Aug 2026 00:00:00 GMT</pubDate><category>Methodology</category></item><item><title>AI as a Partner, Not a Tool</title><link>https://lab.miao-yu.com/en/insights/ai-as-partner/</link><guid isPermaLink="true">https://lab.miao-yu.com/en/insights/ai-as-partner/</guid><description>The dividing line is not how strong the model is — it is whether you give it a chance to disagree with you.</description><pubDate>Sun, 09 Aug 2026 00:00:00 GMT</pubDate><category>Methodology</category></item></channel></rss>