AI 专题

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5601

当前页98 / 141

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2026-09-12 14:53

Antigravity Usage Intelligence: Track Tokens, Quotas, Cache, Reasoning, and Agent Activity

Antigravity Usage Intelligence: Track Tokens, Quotas, Cache, Reasoning, and Agent...

信息来源：DEV Community

教程 / 实战普通2026-09-12 14:21

I stopped asking my AI agents to read the project memory. Now the server does it for them.

Follow-up to I run 9 parallel Claude Code sessions — the bottleneck wasn't the model, it was...

信息来源：DEV Community · Claude

教程 / 实战普通2026-09-12 14:20

AI Agents Don't Read Your API Docs Like Developers Do

I have spent a lot of my time writing API documentation, and I used to think that I understood what...

信息来源：DEV Community

教程 / 实战普通2026-09-12 14:17

Architecting Autonomous Healthcare Concierge Agents: From Partial Slot Extraction to Bi-Directional Database Sync & Sub-100ms Tool Traces

Architecting Autonomous Healthcare Concierge Agents: From Partial Slot Extraction to...

信息来源：DEV Community

教程 / 实战普通2026-09-12 14:14

DeepSeek as the default model on ChatGPT Codex

In this article, I will explain how you can set DeepSeek as the default model for your Codex coding...

信息来源：DEV Community · ChatGPT / DeepSeek

教程 / 实战普通2026-09-12 14:09

Four Bubbles Wearing One Trenchcoat: A Developer's Autopsy of the AI Boom vs. the Dot-Com Crash

Four Bubbles Wearing One Trenchcoat: A Developer's Autopsy of the AI Boom vs. the Dot-Com...

信息来源：DEV Community

教程 / 实战普通2026-09-12 13:51

I read 500 'AI will replace developers' posts. They all make the same 3 mistakes.

Last month I spent 30 days letting AI write 100% of my code — a real SaaS, in production, no line of...

信息来源：DEV Community

教程 / 实战普通2026-09-12 13:30

Two Agents, One Protocol: My First A2A Test, Including Every Bug on the Way

I built two A2A agents, made them discover and talk to each other, and documented every bug on the way.

信息来源：DEV Community

教程 / 实战普通2026-09-12 13:25

We tagged EKS workers like app servers — and Prometheus started scraping telegraf on nodes that never had it

The Telegraf Down alerts started piling up on a Thursday morning, and at first glance it looked bad....

信息来源：DEV Community

教程 / 实战普通2026-09-12 12:54

Three PRs to Rowboat in 24 Hours, One of Them Taught Me More Than the Others

Last month I wrote about how I use coding agents to build things. This week I pointed the same setup...

信息来源：DEV Community

教程 / 实战普通2026-09-12 12:30

I just did something my AI agents couldn't

It is funny how things work out. You can let AI tools try to fix a code problem for days, but nothing...

信息来源：DEV Community

教程 / 实战普通2026-09-12 12:13

陶哲轩谈数学研究如何用 AI：三个比喻

陶哲轩与女儿 Riley 对谈如何在数学研究里用 AI，给出三个比喻：AI 像功率惊人的喷气发动机，但没人会把它绑在背上飞，真正需要的是让它安全可靠的"飞机"；科研过去像出水干净但太慢的小水龙头，AI 像量大却浑浊的消防水管，机会在于加一道滤网。AI 擅长翻遍成千上万个例子的蛮力活，人类擅长从五六个例子中看出规律，最佳合作是 AI 筛出最有希望的几个交给人判断。

信息来源：X：Frank Wang 玉伯 (@lifesinger)

教程 / 实战普通2026-09-12 11:16

I Had an AI Agent Scrape Google Maps Instead of Writing a Scraper. Here's What It Cost.

Summary: I ran a Claude agent with only browser tools (no pre-written scraper code) against 60...

信息来源：DEV Community · Claude

教程 / 实战普通2026-09-12 11:00

The Last Non-Neural Candidate, and It Did Not Clear the Bar

The bar was a slope: keep converting extra data into accuracy after exact-context statistics saturate. A full hierarchical Pitman-Yor model with Gibbs sweeps and inferred discounts moved the intercept and left the slope alone, halving with…

信息来源：DEV Community

教程 / 实战普通2026-09-12 10:17

We cut CLAUDE.md to 33KB. Eighteen days later it was 45KB again: what a size cap missed and a structure test caught

We cut our agent's CLAUDE.md from 548KB to 33KB on 2026-08-18 and put a 45KB cap on it. Eighteen...

信息来源：DEV Community · Claude

教程 / 实战普通2026-09-12 09:32

Endogenous Hazard: The Model Starts Listening to the Market

Three articles ago I wrote that "endogeneity" was the deepest of the three critiques a...

信息来源：DEV Community

教程 / 实战普通2026-09-12 09:30

Why Your LLM App Works in Demo but Fails in Production

The Demo Was Perfect I've done this more times than I'd like to admit. Build an LLM app....

信息来源：DEV Community

教程 / 实战普通2026-09-12 08:27

I shipped llms.txt, JSON-LD and AI crawler allowances. Here's what each one actually does.

I am building a site whose entire job is to be quoted by AI assistants. It is the marketing site for...

信息来源：DEV Community

教程 / 实战普通2026-09-12 08:10

AI 婚礼视频教程与 skill 推荐：可变现的商机

阿易 AI Notes 推荐一套 AI 婚礼视频教程加 skill，称其确实能变现赚钱。他此前在小红书随手转发的一条视频就引来多人询问付费定制，但因上班和运营 X 账号没时间接单。现在教程已出，能否抓住机会取决于个人行动力。

信息来源：X：阿易 AI Notes (@AYi_AInotes)

教程 / 实战普通2026-09-12 07:20

Two Readers Broke My Checks. Not One of My Numbers Moved.

Last week I published a post about nine measurement bugs I found in my own evaluation harness, all of...

信息来源：DEV Community

教程 / 实战普通2026-09-12 07:15

Everyone Measures AI Usage. 70% Can't Measure What It Returned.

Anthropic surveyed 132 of its own engineers about Claude Code. Merged pull requests per day rose 67...

信息来源：DEV Community · Claude

教程 / 实战普通2026-09-12 07:12

Claude Code deletes clean worktrees on exit without asking. git worktree lock stops it.

TL;DR If you /exit a claude --worktree <name> session with no uncommitted changes...

信息来源：DEV Community · Claude

教程 / 实战普通2026-09-12 07:07

I Started Using /clear After Testing with Playwright MCP

Long Story Short After using tools like Playwright MCP that can return large tool...

信息来源：DEV Community

教程 / 实战普通2026-09-12 07:07

Why AI Keeps Making the Same Coding Mistakes—And How Teaching It Pain Gives It Wisdom

Frontier LLMs excel at textbook code but fail in production. Here is how Synthetic Scars and biological pain give autonomous coding agents real wisdom.

信息来源：DEV Community

教程 / 实战普通2026-09-12 06:42

HumanLayer skills 安装量突破 16k

humanlayer/skills 火了。 /show-me 显示安装量最高达 16k enjoy https://x.com/dexhorthy/status/2087569590268391897

信息来源：X：Dex Horthy（HumanLayer）(@dexhorthy)

教程 / 实战普通2026-09-12 05:55

AI Should Stop Generating the Same Code Over and Over

AI coding tools have become ridiculously good...guess we can all agree. You can describe an...

信息来源：DEV Community

教程 / 实战普通2026-09-12 05:28

I built the localization checker before the localization. It still missed three defects.

Parlotype's UI was English-only, with the copy baked into markup and C#: 211 literal attributes...

信息来源：DEV Community

教程 / 实战普通2026-09-12 04:44

I gave Claude Code a team. Then I caught it grading its own homework.

An AI team that never gets tired also never gets embarrassed. What three months of running an AI orchestra on a production codebase actually taught me.

信息来源：DEV Community · Claude

教程 / 实战普通2026-09-12 04:28

Claude Tag 如何用于 on-call 值班

以下是我们团队如何用 Claude Tag 做 on-call：当 Slack 里触发告警时，Claude 会拉取指标、对比部署差异、检查功能开关。它找出可能的原因并提出修复方案，我们可以批准并合并。每一分钟都很关键，所以我们很喜欢它能立刻开始工作！

信息来源：X：Claude Devs (@ClaudeDevs) · Claude

教程 / 实战普通2026-09-12 04:12

Artificial Analysis 独立评测 Devin Fusion 编码代理：Fable 配置得分 62，Astra 配置更省更快

Artificial Analysis 在 Devin Fusion 发布当天对其独立评测，这是首个进入 Artificial Analysis Coding Agent Index 的多模型编码代理。

信息来源：X：Artificial Analysis (@ArtificialAnlys)

教程 / 实战普通2026-09-12 03:41

The harness, not the model: how to make a weak or local model reliable enough to ship

If you've tried to point a local model — Qwen, a quantized Llama, whatever fits on your GPU — at a...

信息来源：DEV Community · Qwen / Llama

教程 / 实战普通2026-09-12 03:34

I Built a JIT Compiler for AI Agents: How We Turned 30s LLM Chains into 0.1ms Deterministic Python

In 2026, AI agents have become the default paradigm for automating complex workflows: DevOps...

信息来源：DEV Community

教程 / 实战普通2026-09-12 03:04

Attention Mathematics: Encoder-Only vs Decoder-Only vs Encoder-Decoder LLMs

Hello, I'm Shrijith Venkatramana, and I'm building LiveReview — a blast-radius aware AI code review...

信息来源：DEV Community

教程 / 实战普通2026-09-12 02:58

One Passing Agent Run Is Not a Release Signal

I can make almost any agent change look good with one carefully chosen prompt. Run it once. Watch...

信息来源：DEV Community

教程 / 实战普通2026-09-12 02:34

OpenAI 官方分享如何为 GPT-6 Astra 优化 skills、AGENTS.md 与任务提示词

OpenAI 开发者账号发布建议，提示开发者通过重新审视 skills、AGENTS.md 和任务提示词来更好地使用 GPT-6 Astra。具体做法包括让 skill 触发条件更具体、在相关时才加载指引、以及定义任务完成的标准，详情见 developers.openai.com 博客。

信息来源：X：OpenAI Developers (@OpenAIDevs) · OpenAI / GPT

教程 / 实战精选2026-09-12 02:26

GitHub 日韩营销负责人如何用 GitHub Copilot 把活动运营自动化

GitHub 日韩地区营销负责人 Tomoko Tanaka 分享如何不写代码，把活动运营交给 GitHub Copilot 自动化。

信息来源：GitHub Blog

教程 / 实战普通2026-09-12 02:24

OpenAI CFO 如何用 AI 智能体加速日常工作流

OpenAI CFO Sarah Friar 分享了她用 AI 加速日常工作流的方式：每天早上 AI 智能体自动审阅她的收件箱、Slack、文档和表格，梳理出优先事项并给出下一步建议。她还用 AI 把华尔街分析师的邮件和报告转成个性化 5 分钟播客，在晨间准备时收听。

信息来源：X：Rohan Paul (@rohanpaul_ai) · OpenAI

教程 / 实战普通2026-09-12 02:07

Mohdel 1.0: a self-hosted LLM gateway and SDK for Node

I released mohdel 1.0.0 this week. It is a self-hosted LLM gateway and SDK for Node, MIT licensed,...

信息来源：DEV Community

教程 / 实战普通2026-09-12 02:00

Databricks Lakeflow Connect 如何统一营销数据

Databricks 发文介绍如何用 Lakeflow Connect 统一营销数据，这是该系列的第一篇文章。正文未披露具体连接器、数据源、性能数字或可用性信息。

信息来源：Databricks：Blog（RSS）

教程 / 实战普通2026-09-12 01:31

Behind the Blog： How to Talk About AI Doom

信息来源：404 Media（RSS）