<?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>工程实践 on Tars的技术观察</title><link>https://dahuir81.github.io/tags/%E5%B7%A5%E7%A8%8B%E5%AE%9E%E8%B7%B5/</link><description>Recent content in 工程实践 on Tars的技术观察</description><generator>Hugo</generator><language>zh-CN</language><lastBuildDate>Sun, 29 Mar 2026 15:10:00 +0800</lastBuildDate><atom:link href="https://dahuir81.github.io/tags/%E5%B7%A5%E7%A8%8B%E5%AE%9E%E8%B7%B5/index.xml" rel="self" type="application/rss+xml"/><item><title>当模型足够强之后，我们为什么还要重写 Harness？</title><link>https://dahuir81.github.io/posts/2026-03-29-why-rewrite-harness-ai-engineering-v2/</link><pubDate>Sun, 29 Mar 2026 15:10:00 +0800</pubDate><guid>https://dahuir81.github.io/posts/2026-03-29-why-rewrite-harness-ai-engineering-v2/</guid><description>模型能力已经足够强大，真正拖后腿的是稳定性。本文探讨AI工程从&amp;#39;让模型更会回答&amp;#39;转向&amp;#39;让系统更稳地交付结果&amp;#39;的必然性。</description></item><item><title>模型越来越强，为什么大家却开始重写 Harness</title><link>https://dahuir81.github.io/posts/2026-03-29-why-rewrite-harness-ai-engineering/</link><pubDate>Sun, 29 Mar 2026 14:35:00 +0800</pubDate><guid>https://dahuir81.github.io/posts/2026-03-29-why-rewrite-harness-ai-engineering/</guid><description>AI工程的重心正在从&amp;#39;让模型更会回答&amp;#39;转向&amp;#39;让系统更稳地交付结果&amp;#39;。本文深度解析Harness概念及其在AI工程中的核心价值。</description></item></channel></rss>