AI 教程与实战
收录 AI 工具教程、提示词、安装部署、配置指南、开发实战与可复现步骤。
分类文章8448
当前页71 / 212
持续更新永久归档
分类文章
按发布时间从新到旧 · 每页 40 条
教程 / 实战普通
What I Look for Before Trusting an AI App Builder With a Real Project
AI app builders can now generate a working database, authentication flow, and deployed URL from a...
信息来源:DEV Community
教程 / 实战普通
InvoicePilot AI: Smarter Invoice Automation with TEv1 and MCP
This is a submission for the Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle ...
信息来源:DEV Community
教程 / 实战普通
Claude Code Git Integration in 2026: How the Agent Reads Diffs, Stages Hunks, and Decides When to Commit
Most AI coding assistants treat Git as an afterthought. Claude Code reads diffs natively, stages hunks interactively, and decides when to commit based on semantic scope. Learn the patterns that make version control safe with an autonomous …
信息来源:DEV Community · Claude
教程 / 实战普通
Building ora-tui with Google Antigravity & Gemini 4 Argon: An MS-DOS Retro TUI & OpenTelemetry Exporter for Oracle RAC, DB & SAN Multipath
How I pair-programmed with Google Antigravity powered by Gemini 4 Argon to build a zero-dependency Go TUI and OpenTelemetry/Prometheus observability stack for Oracle Databases and Linux SAN storage.
信息来源:DEV Community · Gemini
教程 / 实战普通
AI Models Can Learn from Data. But What Can They Learn from Other Models?
Hello, I'm Rijul, and I'm building LiveReview — a blast-radius aware AI code review built for your...
信息来源:DEV Community
教程 / 实战普通
The SEO Score Trap: How to Prove Search Visibility in the Age of AI Answers
A practical, evidence-first guide to technical SEO, AEO, and GEO. Learn how to diagnose real issues, measure AI search visibility, and verify whether your optimizations actually work.
信息来源:DEV Community
教程 / 实战普通
用 Claude Code Ultracode 模式在 SQLite 里实现 Rust 版 SPARQL 1.1
Simon Willison 用 Claude Code for web 的 Ultracode 模式生成 sparqlite,一个用 Rust 编写、把 SPARQL 1.1 嵌入 SQLite 的扩展。
信息来源:Simon Willison 博客 · Claude
教程 / 实战普通
What I Actually Learn When I Build With AI
I've spent entire nights building something with AI, gotten a bunch of features working, and then...
信息来源:DEV Community
教程 / 实战普通
Anthropic 日本 BDR 讲解如何用 Claude Managed Agents 运行六个日常 Agent
Anthropic 日本团队业务拓展代表介绍自己如何把线索分派、线索资格筛选、邮件回复、Salesforce 更新和会议跟进交给一组名为 Nakama 的 Claude Agent,并迁移到 Claude Managed Agents(beta)运行。
信息来源:Claude:Blog(网页) · Claude
教程 / 实战普通
PJ 如何帮小企业上手并自主运行 OpenClaw
工程师出身的 PJ 在佛州 Ocala 经营虫害防治公司期间深度使用 OpenClaw,随后转型为顾问,帮屋顶、泳池、汽车等本地小企业部署 OpenClaw 并培训其自主运行。他的客户不愿把公司数据交给软件、被平台锁定,看重装在办公室电脑里的 OpenClaw 归自己所有。66 岁的行政人员 Donna 每天使用 OpenClaw,称已无法想象没有它的工作。
信息来源:OpenClaw:Blog(网页)
教程 / 实战普通
Vercel Ship:公司构建即智能体构建
公司构建就是智能体构建 ✨ #vercelship 的一大关键收获
信息来源:X:indigo (@indigox)
教程 / 实战普通
Databricks AI 研究团队如何使用 AI Runtime(AIR)
Databricks AI 研究团队披露其作为 AI Runtime(AIR)首位客户的使用方式:AIR 提供 NVIDIA H100 或 B300 GPU,支持从数十亿到近万亿参数模型的 SFT、离线 RL 与在线 RL 训练。
信息来源:Databricks:Blog(RSS) · NVIDIA
教程 / 实战普通
Slop Coding: The Vibe Archeologist Strikes Again!
This is a submission for the Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle ...
信息来源:DEV Community
教程 / 实战普通
Atlassian 如何在 Databricks 上构建 OCSF 安全湖仓
Atlassian 基于 OCSF 开放模式在 Databricks 上构建安全湖仓,处理每天 440 亿条事件。结构化数据配合分区与 Z-ORDER 将代表性查询从约 45 秒降至 2–3 秒,提升 95%;启用 Graviton 集群与 Photon 后查询成本和运行时间减少约三分之二。
信息来源:Databricks:Blog(RSS)
教程 / 实战普通
Worktree lifecycle for AI agents
The problem is not creating worktrees, but managing them When two or more agents work in...
信息来源:DEV Community
教程 / 实战精选
Meta 如何用 AI 原生工程方式开发 Forum 应用
Meta 工程团队详解如何把 AI 智能体作为开发流程参与者来构建独立 Groups 应用 Forum,采用 human-on-the-loop 模式并以验证通过的生产级产出衡量成功。Forum 在八周内达到与 Facebook Groups 的核心功能对齐,自主压测循环在开发期发现并修复了内存问题,内存占用和缺页约减半,上线后智能体还协助根因分析用户反馈。
信息来源:Meta Engineering Blog(RSS)
教程 / 实战普通
LangChain 详解 agent 评估常见问题:从数据集搭建到 Harbor 基准
LangChain 基于 LangSmith 团队的实践经验,回答 agent 评估最常见的入门与进阶问题:agent 输出多样、路径不唯一,传统单元测试难以覆盖,需要从 trace 中抽取 10–20 个示例建数据集、跑基线并持续把线上失败加入数据集形成改进循环。
信息来源:LangChain:Blog(RSS)
教程 / 实战普通
Scope-Creep Guard: A Local Gemma AI That Tells Clients "That's Out of Scope" (So You Don't Have To) published: true
This is a submission for the Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle ...
信息来源:DEV Community
教程 / 实战普通
GitHub 如何用 Evergreen 自动化共享组织的访问审查与仓库清理
GitHub 用原生功能构建的 GitHub App Evergreen,通过两个工作流自动化组织卫生:每季度审查外部协作者访问权限,未响应即默认撤销;检测并归档长期无人类活动的仓库,且从不删除仓库内容。它由 Go webhook 服务器驱动,状态以 JSON 文件提交到 controller repo,用 git 历史充当审计记录,决策通过仓库内的 checklist issue 完成。
信息来源:GitHub Blog
教程 / 实战普通
Google Gemini 演示用 Gemini 4 Argon 在 Canvas 中搭建 3D CAD 工作台制作家具配件
Google Gemini 演示用 Gemini 4 Argon 为家具制作 3D 打印配件。上传带大致尺寸的照片后,可在 Canvas 中构建交互式 3D CAD 工作台,探索不同风格、用滑杆实时调整并导出可打印的 STL 文件。
信息来源:X:Gemini (@GeminiApp) · Gemini
教程 / 实战普通
Eight Minutes to Live. Was It Cursor, or Was It the Spec?
A tip calculator, a stopwatch, and a lot of pesos. TL;DR: I built a USD/MXN tip calculator with...
信息来源:DEV Community · Cursor
教程 / 实战普通
Architecting Autonomous Lakehouse Agents: Governed Iceberg Analytics with Gemini 2.5 Flash and BigLake
The decoupling of compute from storage has positioned Apache Iceberg as the de facto open table...
信息来源:DEV Community · Gemini
教程 / 实战普通
OpenAI Conference Room Bug, Anthropic's Scion vs. Meta's Mononoke, and Cloudflare's Yandex Bot Trace
Inside the $23.99 ESP32 audio recorder found in an OpenAI executive conference room, why Anthropic replaced Git with Scion at 30,000 merges a day, and Cloudflare's JA4 bo
信息来源:DEV Community · OpenAI / Claude
教程 / 实战普通
AI-Generated Tests Can Agree With the Bug
When an AI agent writes both the implementation and its tests, a green build may only prove that the same wrong assumption was encoded twice.
信息来源:DEV Community
教程 / 实战普通
Local image generation with llamadart for the Llama Village storybook
Beyond chat with llamadart · Part 4 Dash missed it. Pip’s scarf is by the pond, and all he has is...
信息来源:DEV Community · Llama
教程 / 实战普通
Introducing B-TAP: Because Your AI Infrastructure Hasn't Attributed Its Fair Share
Context isn't free. Retrieval isn't free. Observability isn't free. And an international standards...
信息来源:DEV Community
教程 / 实战普通
RevCheck, compare the cars you want before you buy your next daily driver
This is a submission for the Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle ...
信息来源:DEV Community
教程 / 实战普通
It caught "partner" "pareja" in an IRS rule and missed "return" "devolución": an offline MT checker for translators
An offline checker that ranks machine-translated segments by the risk of a major meaning error: a $1.22 Tinker LoRA, Gemma 4 on DigitalOcean, Sentry tracing.
信息来源:DEV Community
教程 / 实战精选
Claude Code 团队如何做代码审查:六层检查与风险分级审批
Anthropic Claude Code 团队详解在 Claude 写约 90% 代码的情况下如何组织代码审查:审查从终点关卡变为持续进行,每个 PR 经过验证、必需 CI、AI 审查、安全审查、按风险分级的人工审批和合并队列六层检查,再辅以快速回滚。
信息来源:Anthropic:Claude.dev 开发者博客(RSS) · Claude
教程 / 实战普通
Your AI Agent Just Got a Passport: Agent Identity on Google Cloud and the Redeploy Trap Nobody Talks About
Why your agents need their own identity Most teams run every AI agent under one shared...
信息来源:DEV Community
教程 / 实战普通
Kimi Work 工作流自动化教程第 3 期
用 Kimi Work 自动化你的工作流 - 教程 #3 看看 Kimi 如何融入你的日常工作流: - 跨本地项目、浏览器和电脑工作。 - 用 Swarm 拆解复杂任务;在 Goal 模式下迭代。 - 通过 Plugins、AGENTS.md 和 Dashboard 自定义 Kimi 的工作方式。 接下来想让我们讲什么?在评论区告诉我们!
信息来源:X:Kimi.ai (@Kimi_Moonshot) · Kimi
教程 / 实战普通
State of AI Instruction Quality: A Third of Repos Shout
IMPORTANT. CRITICAL. You MUST. NEVER. A third of the repos in the Q3 report, 3,319 of the 9,784 with...
信息来源:DEV Community
教程 / 实战普通
ScopeGuard: an open-weight model that tells freelancers when "quick one" means "paid one"
This is a submission for the Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle ...
信息来源:DEV Community
教程 / 实战普通
I built a lie detector for AI coding agents. Then I read the rules.
When an AI agent commits a change to code, it makes a claim about what it changed: "fixed typos" or...
信息来源:DEV Community
教程 / 实战普通
卡兹克用 Claude 搭建公司中台定时任务并分享省 Token 技巧
作者卡兹克搭建了公司中台的 Claude 自动化流程,定时任务每10分钟检查后台是否有新意见,Claude 会在后台启动并根据需求判断是否开发优化、改完后回复。省额度技巧是平时用 Haiku 5.5 探测有无新意见,有的话再启动 Opus 5.5 子Agent 来解决,以最大化 Token 利用。
信息来源:X:卡兹克 (@Khazix0918) · Claude
教程 / 实战普通
I Vibe-Coded My Way Out of Angular Upgrades
Do you also hate upgrading Angular versions? Because I do. I maintain a few open-source projects, and...
信息来源:DEV Community
教程 / 实战普通
LangChain 团队以 Q&A 形式分享 Agent Evals 常见问题实践指南
邵猛转发 LangChain 团队 Divya Santhanam 以 Q&A 形式发布的 Agent Evals 实践指南(https://www.langchain.com/blog/common-questions-about-agent-evals)。
信息来源:X:邵猛 (@shao__meng)
教程 / 实战普通
/grill-me Skill 拟加 GRILL.md 配置文件
Matt Pocock 提议为 /grill-me Skill 增加 GRILL.md 配置文件,用于定制拷问式提问的行为。它支持用配置组合取代命令变体,如在 GRILL.md 中写"调用 domain-modeling Skill"即可替代 /grill-with-docs;还能定制规划阶段关注的侧重点(如安全、模块划分策略),以及选择逐条提问还是一次性列出全部问题。
信息来源:X:邵猛 (@shao__meng)
教程 / 实战普通
Simon Willison 用 Claude Code 构建 LaTeX 渲染工具,文档存于 URL fragment
Simon Willison 受 Tiny Little Office 启发,用 Claude Code(Opus 5.5 Extra)构建了一个 LaTeX 渲染工具,通过 URL 的 #fragment 存放 LaTeX 代码并渲染,支持自定义压缩字典。
信息来源:Simon Willison 博客 · Claude
教程 / 实战普通
Gemma 4's MTP Drafter, Explained: Speculative Decoding and a 4-Bit Drafter at 60% of the Size
What speculative decoding is, how Gemma 4's multi-token prediction drafter rides on the main model, why you would add one, and an exact Q4_0 repack of Google's QAT drafters on Hugging Face that cuts the file to 60% of the size and the memo…
信息来源:DEV Community · Hugging Face