<?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>LLM工程 on 硅基 AGI · 智能体学习与测评</title><link>https://guijiagi.com/tags/llm%E5%B7%A5%E7%A8%8B/</link><description>Recent content in LLM工程 on 硅基 AGI · 智能体学习与测评</description><generator>Hugo</generator><language>zh-cn</language><copyright>本站内容采用 CC BY-NC-SA 4.0 国际许可协议授权</copyright><lastBuildDate>Thu, 02 Jul 2026 11:34:00 +0800</lastBuildDate><atom:link href="https://guijiagi.com/tags/llm%E5%B7%A5%E7%A8%8B/index.xml" rel="self" type="application/rss+xml"/><item><title>AI成本优化2026实战</title><link>https://guijiagi.com/posts/ai-cost-optimization-2026/</link><pubDate>Thu, 02 Jul 2026 11:34:00 +0800</pubDate><guid>https://guijiagi.com/posts/ai-cost-optimization-2026/</guid><description>从模型选择到缓存策略，AI系统成本优化的全方位实战指南</description></item><item><title>RAG还是微调：决策框架</title><link>https://guijiagi.com/posts/rag-vs-fine-tuning-decision/</link><pubDate>Thu, 02 Jul 2026 11:31:00 +0800</pubDate><guid>https://guijiagi.com/posts/rag-vs-fine-tuning-decision/</guid><description>何时用RAG、何时用微调？基于多维度评估的系统化决策框架</description></item><item><title>AI系统测试策略</title><link>https://guijiagi.com/posts/ai-testing-strategies/</link><pubDate>Thu, 02 Jul 2026 11:27:00 +0800</pubDate><guid>https://guijiagi.com/posts/ai-testing-strategies/</guid><description>从单元测试到红队测试，AI系统全链路测试策略与实践</description></item><item><title>提示词版本管理：用Git管理</title><link>https://guijiagi.com/posts/prompt-versioning-git/</link><pubDate>Thu, 02 Jul 2026 11:26:00 +0800</pubDate><guid>https://guijiagi.com/posts/prompt-versioning-git/</guid><description>将提示词纳入版本管理，实现提示词的迭代追踪、A/B测试与回滚</description></item><item><title>LLM结构化输出指南</title><link>https://guijiagi.com/posts/structured-output-llm-guide/</link><pubDate>Thu, 02 Jul 2026 11:20:00 +0800</pubDate><guid>https://guijiagi.com/posts/structured-output-llm-guide/</guid><description>从JSON Schema到Constrained Decoding，确保LLM输出结构化数据的完整方案</description></item></channel></rss>