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8439

当前页193 / 211

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2026-09-03 07:17

BestBlogs 早报 09-03：Gemini 3.8 Flash 升级，Anthropic 电商智能体指南与 GitHub Copilot 降本实践

本期早报围绕能力、工作过程与最终成本展开，精讲三篇：Google 发布 Gemini 3.8 Flash 系列含通用 Flash 与 Flash Cyber，内部漏洞发现成功率超 70%。

信息来源：X：洪明 (@hongming731) · Claude / Gemini

教程 / 实战普通2026-09-03 07:11

Claude Code journal plugin: Notion session summaries at a glance

The /journal skill from my breadcrumbs post is now a Claude Code plugin anyone can install. One glanceable callout per topic, colored by season with an icon for the weekday, three Notion calls per run. What it writes, how to set it up, and…

信息来源：DEV Community · Claude

教程 / 实战普通2026-09-03 06:58

RL 1: Biological foundations and the "Law of Effect" (1898– 1949)

Before GPUs. Before neural networks. Before transistors were even a rumor. There was a hungry cat in...

信息来源：DEV Community

教程 / 实战普通2026-09-03 05:51

Making three years of a Telegram group chat queryable

Telegram exports two different shapes, and neither is the hard part. The hard part is packing messages under a 500k-word source cap and making the second run idempotent.

信息来源：DEV Community

教程 / 实战普通2026-09-03 05:30

My AI agent built a flight recorder for AI agents, and it flagged itself

Every developer I know now runs an AI coding agent in something like auto-accept mode. Claude Code,...

信息来源：DEV Community · Claude

教程 / 实战普通2026-09-03 05:16

Alexandr Wang 称 Muse Spark 1.3 的编码智能体成绩被低估并给出 muse code 安装命令

信息来源：X：Alexandr Wang（Scale AI 创始人/Meta 首席 AI 官） (@alexandr_wang)

教程 / 实战普通2026-09-03 04:41

Why "tests passed" is the wrong question for an AI coding agent

Your coding agent finishes a task and tells you the tests passed. You accept the report. Later, when...

信息来源：DEV Community

教程 / 实战普通2026-09-03 03:29

How Does AI Know What's Inside an Image? Meet Vision Language 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2026-09-03 03:16

How to stop your AI from finding new errors on every review

A review protocol that challenges the scheme first, then the code, then stops. Full prompt included.

信息来源：DEV Community

教程 / 实战普通2026-09-03 03:07

Claude Code 团队如何使用 Claude Code

Claude 官方频道发布视频，介绍 Claude Code 团队自己如何使用 Claude Code。材料仅含标题，无正文文本，具体内容细节以视频为准。

信息来源：Claude：YouTube（RSS） · Claude

教程 / 实战普通2026-09-03 02:37

Three registers, three alphabets, no APIs: building pay-per-event actors for company data

I build and sell Reestri, a small catalog of paid Apify actors that answer one kind of question: who...

信息来源：DEV Community

教程 / 实战精选2026-09-03 02:00

GitHub Copilot 如何在不牺牲任务质量的前提下降低 AI 编码成本

GitHub 工程师 Erik Kristensen 分享了 Copilot 降本的四项改动：选择性压缩工具输出、移除 view 工具行号前缀（线下推理成本降约 5%，线上用户日均推理成本降约 3%）、压缩 task-tool 提示词（每轮省约 1300 token，每活跃小时归一化成本降 2.9%）、后台任务完成后直接交付结果（AI Credits 用量降约 2.3%）。

信息来源：GitHub Blog

教程 / 实战普通2026-09-03 01:01

Beyond the Demo: Architecting a Security-First AI Journal with Gemini

Most "AI-powered" side projects follow the same arc: a slick demo, a hardcoded API key, one shared...

信息来源：DEV Community · Gemini

教程 / 实战普通2026-09-03 00:57

🌟 The CI Gate Rejected the Terraform Change—but the LLM Still Ran

How I added a deterministic trace contract to an AI Terraform reviewer so rejected infrastructure...

信息来源：DEV Community

教程 / 实战普通2026-09-03 00:57

I Let AI Design Tools Build My Portfolio Site — Clients Noticed

A portfolio website should make it easy for a potential client to answer three questions: What do you...

信息来源：DEV Community

教程 / 实战普通2026-09-03 00:54

The Best Academic Search API for Humans and AI Agents in 2026

Quick answer: the best academic search API for AI agents in 2026 is Valyu, because it is the only one...

信息来源：DEV Community

教程 / 实战普通2026-09-03 00:38

What do you build when you can build anything?

My hot take is: Telling people that they have to build endlessly is very, very, very bad advice. If...

信息来源：DEV Community

教程 / 实战精选2026-09-03 00:35

Google AI 团队分享如何为 LLM-as-a-Judge 评测编写可靠的评分标准

Google AI 团队发布教程，讲解如何为 LLM-as-a-Judge 评测编写可靠的布尔式评分标准，指出模糊提示会导致评估不一致和浪费 token。文中给出四条经验：问题保持原子化且互不重叠、只让评判模型评估客观事实（可用 RFC 2119 术语如 MUST 表述）、只评 prompt 中明确要求的内容、用专家标注的 golden set 校准评判模型直至与人类评分一致。

信息来源：Google AI：DEV 作者专属（RSS）

教程 / 实战普通2026-09-03 00:21

My AI Gateway Added 400ms to Every Request. Here's Where It Went

If your AI application suddenly becomes 300–500ms slower after adding an AI gateway, the first...

信息来源：DEV Community

教程 / 实战普通2026-09-03 00:16

Why I'm Building ShrekOS When Containers Already Exist

Useful AI agents keep forcing a bad choice: lock the environment down until it is useless, or hand the agent your real machine. ShrekOS is my attempt to move that trust decision below the agent, into the operating system.

信息来源：DEV Community

教程 / 实战普通2026-09-03 00:07

Your System Prompt Has a Shelf Life: Maintaining Prompts as Models Improve

Anthropic removed over 80% of Claude Code's system prompt for Claude Opus 5 and Claude Fable 5, and...

信息来源：DEV Community · Claude

教程 / 实战普通2026-09-02 23:52

赵昊与百度百舸团队对谈：世界模型如何走向规模化训练

百度智能云具身智能前沿讲堂上，清华大学AIR助理教授赵昊与百度智能云AI计算首席科学家王雁鹏等对谈世界模型规模化问题。赵昊提出以3D/4D Token为核心的世界模型路线，认为世界模型不等于视频生成，需表达空间结构、光学属性和速度、加速度等动态信息；其OmniNWM通过动作编码与相机参数解耦，实现200帧稳定生成，并与精锋医疗、极智嘉分别合作手术预演和仓库机器人场景。

信息来源：公众号：百度智能云（文心）

教程 / 实战普通2026-09-02 23:40

React, Next.js, Svelte, Zod: none of them can tell AI who they're actually for

Your coding agent is good at reading code. Point Claude Code or Cursor at a repo and it...

信息来源：DEV Community · Claude / Cursor

教程 / 实战普通2026-09-02 23:15

Shopify's agent-commerce category filter doesn't filter. We checked 190 stores.

Since 2026 every Shopify storefront answers an agent-commerce endpoint at POST /api/ucp/mcp,...

信息来源：DEV Community

教程 / 实战普通2026-09-02 22:29

Every 播客文字稿：明星写手如何把职业教练变成 AI 员工

Every 发布播客文字稿，讲述其明星写手把自己的职业教练改造成 AI 员工的经历。正文无有效文本，具体方法与细节以原文为准。

信息来源：Every：最新文章（网页）

教程 / 实战普通2026-09-02 22:15

Elvis Saravia：AI 工程师应从零构建最小 harness 掌握上下文工程

Elvis Saravia 提出 harness 并非千篇一律，好的 harness 应解决上下文工程问题，关键在于理解并优化系统提示词、缓存和有效工具调用等底层细节。他建议用自己喜欢的语言构建一个带基本 agent 循环、系统提示词和工具调用能力的最小 harness，在用例上测试验证并迭代，掌握这些可调旋钮后再迁移到生产级 harness。

信息来源：X：Elvis Saravia (@omarsar0, DAIR.AI)

教程 / 实战普通2026-09-02 22:04

Ethan Mollick 推荐 AI 时代之前的三页复杂系统失效论述，并提出用 AI 构建系统防御新哲学

Ethan Mollick 分享 Richard I. Cook 写于 AI 出现之前的《How Complex Systems Fail》三页内容，称其值得阅读。核心观点是复杂系统带病运行但仍存活，因为缺陷很少同时对齐，给了人类介入的机会；而 AI 能发现或制造对齐的缺陷，因此需要一套借助 AI 的系统防御新哲学。

信息来源：X：Ethan Mollick (@emollick)

教程 / 实战普通2026-09-02 21:54

From Local Traces to Production Observability for Google AI Agents

Most difficult agent incidents begin with one question: Why did the agent do that? Why did it...

信息来源：DEV Community

教程 / 实战普通2026-09-02 21:19

I Found 3 Security Vulnerabilities in My Own AI Agent's Tool Access

I built GeoMart for OpenAI's WebMCP Challenge: a storefront where a human fills in a live,...

信息来源：DEV Community · OpenAI

教程 / 实战普通2026-09-02 21:18

The Last Time I Googled an Error Message

I don't remember the last time I pasted an error into Google. Not the date. I mean the habit left...

信息来源：DEV Community

教程 / 实战普通2026-09-02 21:03

玉伯谈 Codex 用户卡点：安装是第一关，用来做什么是最后一关

Frank Wang 玉伯发帖称，大量用户的第一个卡点不是怎么用 Codex，而是怎么安装 Codex。他表示用户在安装好、解决第一个卡点后，马上会遇到最后一个卡点，即用 Codex 来做什么，然后就没有然后了。

信息来源：X：Frank Wang 玉伯 (@lifesinger)

教程 / 实战普通2026-09-02 20:46

Stratagems #27: Alex Watched the AI Wait. He Said Everything Was Normal.

Rather pretend not to know than pretend to know and act on it. Stay still. Hide the move. Wait like...

信息来源：DEV Community

教程 / 实战普通2026-09-02 20:33

Accountable Generation: Intent as the Artifact of Record

Every serious engineering organization has spent decades building governance around one artifact:...

信息来源：DEV Community

教程 / 实战普通2026-09-02 20:26

I think I found my ideal job

Have you ever read a job posting and your heart just sings the song of "HIRE ME PLEASE!!!", yeah,...

信息来源：DEV Community

教程 / 实战普通2026-09-02 20:19

GLM-5.3-Flash: A Geopolitical Disruptor Hiding in an Ultra-Cheap Model

Before the world knew it was a product of Z.ai, a massive Chinese tech player, this model was running...

信息来源：DEV Community · GLM

教程 / 实战普通2026-09-02 20:17

My Code Works. An AI Wrote It. Neither of Us Knows Why.

here's a meme we've all posted some version of. "My code works and I have no idea why." Classic....

信息来源：DEV Community

教程 / 实战普通2026-09-02 20:08

73% of What We Blocked in a Month Was One Syscall Your Sandbox Cannot See

For the last month, every AI coding session on this machine ran under grith, which intercepts every...

信息来源：DEV Community

教程 / 实战普通2026-09-02 20:05

I Let an AI Chatbot Answer Buyer Questions. It Almost Cost a Deal

A real estate website can generate a buyer inquiry in seconds, but one wrong chatbot answer can undo...

信息来源：DEV Community

教程 / 实战普通2026-09-02 20:03

I'm 12. A dev asked me to expand my architecture. Here's the deep dive — schema, RLS, fallback chain, and the z-index war.

My last article shared the high-level architecture of KODA (my AI coding mentor built on a POCO C55)....

信息来源：DEV Community

教程 / 实战普通2026-09-02 19:36

How to Stop Leaking API Keys and PII to LLMs: Zero-Trust In-Memory Sanitization for Cursor & Claude

If you are using AI coding assistants like Cursor, Windsurf, Claude Desktop, or ChatGPT, you have...

信息来源：DEV Community · ChatGPT / Claude / Cursor