<?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>AI on Plbear | Cuz Penguin QQ</title><link>https://www.plbear.com/tags/ai/</link><description>Recent content in AI on Plbear | Cuz Penguin QQ</description><generator>Hugo</generator><language>zh-cn</language><lastBuildDate>Mon, 10 Aug 2026 10:00:00 +0800</lastBuildDate><atom:link href="https://www.plbear.com/tags/ai/index.xml" rel="self" type="application/rss+xml"/><item><title>App UI自动化测试实践：基于元素Skills的自动化测试框架设计</title><link>https://www.plbear.com/posts/2026-08-10-app-ui-testing-skills/</link><pubDate>Mon, 10 Aug 2026 10:00:00 +0800</pubDate><guid>https://www.plbear.com/posts/2026-08-10-app-ui-testing-skills/</guid><description>App UI 自动化测试框架的设计与实践，基于元素 Skills 概念，用 AI 辅助元素定位和测试用例生成，解决传统 UI 自动化维护成本高、元素定位不稳定的痛点。</description></item><item><title>高峰论坛活动系统 2.0：AI 生成议程与智能推荐的技术实现</title><link>https://www.plbear.com/posts/2025-09-01-summit-system-20/</link><pubDate>Mon, 01 Sep 2025 14:00:00 +0800</pubDate><guid>https://www.plbear.com/posts/2025-09-01-summit-system-20/</guid><description>高峰论坛活动系统 2.0 的技术实现，AI 自动生成活动议程、智能推荐参会者和感兴趣的议题、实时互动问答，提升活动组织效率和参会体验。</description></item><item><title>职位搜索算法优化：从 BM25 到向量检索的混合召回方案</title><link>https://www.plbear.com/posts/2025-07-23-job-search-algorithm/</link><pubDate>Wed, 23 Jul 2025 10:00:00 +0800</pubDate><guid>https://www.plbear.com/posts/2025-07-23-job-search-algorithm/</guid><description>职位搜索算法从传统 BM25 到向量检索的混合召回方案优化，涵盖语义匹配、向量召回、重排序，解决同义词、语义不匹配等传统搜索痛点，搜索点击率提升 35%。</description></item><item><title>AI 模拟面试智能体架构：Go SSE 流式分发与 Prompt 状态机工程</title><link>https://www.plbear.com/posts/2025-06-10-ai-interview-agent/</link><pubDate>Tue, 10 Jun 2025 14:00:00 +0800</pubDate><guid>https://www.plbear.com/posts/2025-06-10-ai-interview-agent/</guid><description>AI 模拟面试智能体的架构设计，基于 Go SSE 流式分发实现实时对话，用 Prompt 状态机约束面试流程，解决长对话模型失控、流式延迟高、计费闭环等问题。</description></item><item><title>MCP 协议入门与实战：基于 FastMCP 构建企业级 Skills 服务</title><link>https://www.plbear.com/posts/2025-04-21-mcp-fastmcp-practice/</link><pubDate>Mon, 21 Apr 2025 10:00:00 +0800</pubDate><guid>https://www.plbear.com/posts/2025-04-21-mcp-fastmcp-practice/</guid><description>MCP（Model Context Protocol）协议入门与实战，基于 FastMCP 框架构建企业级 Skills 服务，让大模型安全调用招聘平台的结构化数据，实现 AI 智能体的工具调用能力。</description></item><item><title>AI 网关架构设计：One-API 大模型网关的鉴权、限流与计费闭环</title><link>https://www.plbear.com/posts/2025-03-31-ai-gateway-architecture/</link><pubDate>Mon, 31 Mar 2025 14:00:00 +0800</pubDate><guid>https://www.plbear.com/posts/2025-03-31-ai-gateway-architecture/</guid><description>基于 One-API 的大模型网关架构设计，涵盖多模型统一接入、API Key 鉴权、限流策略、Token 计费、会员体系，实现 AI 能力的商业化闭环。</description></item></channel></rss>