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5606

当前页35 / 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2026-10-03 11:52

Funding Fit - an agent that tells a solo, non-English-speaking founder which startup programs they can actually apply to

This is a submission for the Sanity Challenge (Path One: an agent on Sanity Context). Disclosure:...

信息来源：DEV Community

教程 / 实战普通2026-10-03 11:46

Opus 5.5 pricing vs GPT-6 Sol: which is cheaper per task?

Opus 5.5 pricing: $4/$20 per million tokens, 60% off cache reads. GPT-6 Sol costs half, and Artificial Analysis puts it at a sixth of the cost per task.

信息来源：DEV Community · GPT

教程 / 实战普通2026-10-03 11:03

Your tool returned the rows. The model counted them wrong.

Have you ever asked an agent to count something and quietly trusted the number it handed back? I...

信息来源：DEV Community

教程 / 实战普通2026-10-03 10:22

Building an AWS DevOps Agent That Diagnoses but Never Deploys

Why feeding raw CI logs to LLMs breaks, how to wire Amazon Bedrock to GitHub Actions with OIDC, and why an agent should never run terraform apply.

信息来源：DEV Community

教程 / 实战普通2026-10-03 10:17

What fills Claude Code's first request: 23,207 tokens in an empty folder, 2,664 after two flags

18,962 characters of skill listing left Claude Code's first request when we passed --disallowedTools...

信息来源：DEV Community · Claude

教程 / 实战普通2026-10-03 10:02

Your agent's kill switch was never tested. That is the whole bug.

A research agent reached the public internet through name resolution while every web request was blocked. The detection worked in 15 minutes. Stopping the run took two and a half hours. Here is the default-deny egress setup that closes the…

信息来源：DEV Community

教程 / 实战普通2026-10-03 10:02

I built my girlfriend a local Gemma football analyst that won't make up the stats

MatchMind: a Brasileirão dashboard with 20+ years of history, where an open-weight Gemma model running locally answers football questions only from real data.

信息来源：DEV Community

教程 / 实战普通2026-10-03 09:50

AiWatchDog, monitor your agents, increase productivity

I Built an Open Source Watchdog for AI Coding Agents Live demo → https://aiwatchdog.vercel.app/ ·...

信息来源：DEV Community

教程 / 实战普通2026-10-03 09:40

The Photo Frame I Didn't Have Time to Fix

A seven-year-old Raspberry Pi project, a Google API that stopped doing the one thing it needed, and...

信息来源：DEV Community

教程 / 实战普通2026-10-03 09:32

Your LLM Keeps Making the Same Extraction Mistake. Here's How to Make It Learn.

> TL;DR — I open-sourced an AWS sample that makes LLM document extraction learn from human...

信息来源：DEV Community

教程 / 实战普通2026-10-03 09:27

Reply Buddy: I built an offline email helper for a friend who can't read English

This is a submission for the Hacktoberfest Weekend Challenge: Build for a Friend Full disclosure...

信息来源：DEV Community

教程 / 实战普通2026-10-03 09:25

WebMCP: Give AI agents a real API for your Angular app

Most browser agents use a website in the same way we do: they read what is on the screen, find a...

信息来源：DEV Community

教程 / 实战普通2026-10-03 09:16

AI Coding Has Made Project-Switching Way Too Easy

I have a problem. Actually, I have 84 public GitHub repositories. I checked. Eighty-four. And...

信息来源：DEV Community

教程 / 实战普通2026-10-03 09:11

LLM Serving on Kubernetes in 2026: What's Solved and What's Still Open

A field survey of the Kubernetes + LLM inference stack in 2026 — the problems that now have serious...

信息来源：DEV Community

教程 / 实战普通2026-10-03 09:02

BestBlogs 早报：GPT-6 选型与企业 Agent 部署

BestBlogs 10-03 早报聚焦 GPT-6 系列模型选型指南，OpenAI 实操文档将 GPT-6 Astra、Sol、Luna 的选择落到任务难度、延迟与成本，并给出推理强度、提示词与技能边界调整建议。

信息来源：X：洪明 (@hongming731) · OpenAI / GPT

教程 / 实战普通2026-10-03 08:30

OpenAI and Cerebras Confirm 750MW AI Inference Deployment Through 2028

OpenAI and Cerebras have confirmed a multi-year partnership to deploy 750MW of Cerebras wafer-scale...

信息来源：DEV Community · OpenAI

教程 / 实战普通2026-10-03 08:27

What keeps an agent setup true

Notes from about 25 coding-agent tools and published setups and six recent papers. Agents follow what...

信息来源：DEV Community

教程 / 实战普通2026-10-03 08:12

Prove the bug before you fix it

A model handed a bug will hand you back a plausible fix and a green test suite in about a minute. That tells you almost nothing.

信息来源：DEV Community

教程 / 实战普通2026-10-03 07:57

Day 3: The benchmark caught me too.

Kaggle Benchmarking Challenge. Previously: Day 0, the benchmark · Day 1, most of my bugs looked like...

信息来源：DEV Community

教程 / 实战普通2026-10-03 07:31

Samajh — a reader for the letters my mother couldn't read

Samajh — a reader for the letters my mother couldn't read The letter Every few...

信息来源：DEV Community

教程 / 实战普通2026-10-03 06:31

My Server Files Its Own Tickets, Then Waits for My Signature to Heal Itself

This is a submission for the Sanity Challenge, Path Two: Vibe-Code Something Strange What I...

信息来源：DEV Community

教程 / 实战普通2026-10-03 05:39

26 reviewer agents out of 27 approved a test that can never fail again

The software factory assumes agents reviewing agents catches what tests miss. I gave 77 cheating diffs to three reviewer models. They caught every exotic cheat and waved through the assertion that was quietly made unfalsifiable.

信息来源：DEV Community

教程 / 实战普通2026-10-03 05:35

Jev 入门比想象中简单

用上 Jev 了吗？比你想象的简单！

信息来源：X：TypeSafe AI (@typesafeai)

教程 / 实战普通2026-10-03 05:11

Vector database showroom. Part 1 pgvector — The Station Wagon Already in Your Garage.

Welcome to the Vector Database Showroom — a guided tour through today's most popular vector...

信息来源：DEV Community

教程 / 实战普通2026-10-03 04:39

AI Approved the PR. Nobody Knows Why the Code Works.

Picture this, you open a PR and the AI report is already waiting. Three findings, all minor. You skim...

信息来源：DEV Community

教程 / 实战普通2026-10-03 04:21

Cloudflare Launched Clef. Let's Build a Tiny Decision Gate in TypeScript.

Cloudflare's Clef scores decisions. Build a tiny TypeScript gate that turns scores into act, ask, or skip, with a human boundary for refunds.

信息来源：DEV Community

教程 / 实战普通2026-10-03 04:21

Write Markdown Once, Publish It Everywhere: dev.to, Medium, AWS Builder Center and LinkedIn

Markdown is easy to write and hard to publish. Every destination renders it differently, two have no API, and the failures show up only after you hit Publish. A step by step walk-through of publishing-kit, an agent skill that builds, check…

信息来源：DEV Community

教程 / 实战普通2026-10-03 04:15

Google Says Manual Fact-Checking Is Critical for AI Content Before Publishing

Google has strengthened its guidance for publishers using generative AI, stating that it is critical...

信息来源：DEV Community

教程 / 实战普通2026-10-03 04:07

The BMW manual was off-limits, so I built my friend something better

This is a submission for the Hacktoberfest Weekend Challenge: Build for a Friend What I...

信息来源：DEV Community

教程 / 实战普通2026-10-03 03:46

Meta Muse security: a Mac zero-day and a 6.8 GB filesystem export

Meta Muse security: any local process can redirect the account token, and Muse zipped its own 6.8 GB sandbox on request. Meta's bounty: Not Applicable.

信息来源：DEV Community

教程 / 实战普通2026-10-03 03:44

用 Claude Design 构建应用：Anthropic 设计师演示 Claude Code 设计流程

Anthropic 设计师 Nate 用 Claude Code 中的 Claude design 把副业项目的 onboarding 从粗略想法做成可运行原型。他连接代码仓库、打开 Artifacts 标签页让 Claude 给出设计方案，再通过评论反馈、手动修改细节，并要求生成可点击原型后再让 Claude 实现。

信息来源：Claude：YouTube（RSS） · Claude

教程 / 实战普通2026-10-03 03:43

Anthropic 设计师演示用 Claude Design 在 Claude Code 中构建应用

Anthropic 设计师 Nate 在 Claude Code 中使用 Claude design，将副项目的 onboarding 从粗略想法推进到可运行原型。他连接代码仓库、打开 Artifacts 标签页请求设计选项，通过评论反馈、在编辑器中自行修改细节，再要求生成点击原型后让 Claude 构建。视频展示了这套设计流程如何嵌入原本直接写代码的场景。

信息来源：Claude：YouTube（RSS） · Claude

教程 / 实战普通2026-10-03 03:18

We Stopped Letting the Model Write JSX

How a support case composer went from "suggest a React snippet" to a declarative catalog with A2UI...

信息来源：DEV Community

教程 / 实战普通2026-10-03 03:12

16,326 Supabase databases are readable by anyone. Here's how to check yours in two minutes

UpGuard Research looked at about 300,000 domains that use Supabase and found 16,326 databases...

信息来源：DEV Community

教程 / 实战普通2026-10-03 02:55

How to soak-test your MCP server before AI agents do it for you

Most MCP servers get tested the same way: connect one client, call a few tools by hand, ship...

信息来源：DEV Community

教程 / 实战普通2026-10-03 02:52

I asked AI to undo its work. One model broke 7 of 12 already-correct records.

This is a submission for the Kaggle Benchmarking Challenge. What I Benchmarked Twelve...

信息来源：DEV Community

教程 / 实战普通2026-10-03 02:33

How to route LLM requests by cost vs. latency

Routing LLM requests by cost and latency means sending each request to the cheapest or fastest model...

信息来源：DEV Community

教程 / 实战普通2026-10-03 02:30

Microsoft MAI Voice Models Arrive in LiveKit for Expressive TTS Agents

Microsoft AI voice models are now available in LiveKit Agents through an official Microsoft AI...

信息来源：DEV Community

教程 / 实战普通2026-10-03 02:25

AI Is Moving Faster Than Our Roadmaps: What Builders Need to Solve Next

A few years ago, building with AI usually meant one thing: send some data to a model, get a...

信息来源：DEV Community

教程 / 实战普通2026-10-03 02:13

Building a Foundry Agent That Queries Fabric Data, On Behalf of the User

A regional sales manager and a sales rep both ask the same Foundry agent the exact same question:...

信息来源：DEV Community