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5614

当前页53 / 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2026-09-27 07:35

Permission Slip: my AI agents need a signed slip before they do anything public

This is a submission for the Sanity Challenge, Path Two: Vibe-Code Something Strange How this was...

信息来源：DEV Community

教程 / 实战普通2026-09-27 07:00

ClusterMAX 3.0：集群配置如何影响实际性能

在 ClusterMAX 3.0 文章中，我们分享了一些真实案例，展示集群配置的变化如何影响开箱即用的实际性能。最简单的演示方式是用训练工作负载。(1/3)🧵

信息来源：X：SemiAnalysis (@SemiAnalysis_)

教程 / 实战普通2026-09-27 06:44

Don't put your agent guardrails in the system prompt

Coding agents write, refactor, and open pull requests at a pace humans can't match. The instinct is...

信息来源：DEV Community

教程 / 实战普通2026-09-27 06:41

Will It Stack? An agent that knows which Raspberry Pi HATs fight over the same pins

This is a submission for the Sanity Challenge, Path One: Ship an Agent That Queries Real...

信息来源：DEV Community

教程 / 实战普通2026-09-27 06:14

MCP vs. A2A: why treating every agent like a tool is slowing your system down (and how to split the two) — MCP vs. A2A

I spent three weeks trying to fix a latency problem that didn't exist. Our multi-agent customer...

信息来源：DEV Community

教程 / 实战普通2026-09-27 06:13

I’ve been researching a problem that sits underneath almost every autonomous-agent system:

A model being intelligent does not make the agent trustworthy. Reasoning, planning, memory, tools,...

信息来源：DEV Community

教程 / 实战普通2026-09-27 06:07

Stop hand-tuning prompts: build and optimize an LLM program with DSPy

Originally published at AI Frontier Post. Every production LLM feature eventually hits the same...

信息来源：DEV Community

教程 / 实战普通2026-09-27 04:40

I built an async, self-improving AI coding agent that never leaves my machine. Now it runs in my browser, on my phone, and learns from its own work.

Flash 0.5.5 turns a local-first terminal agent into something you can use from any screen in your house, without sending a single line of code to someone else's server.

信息来源：DEV Community

教程 / 实战普通2026-09-27 04:34

Building an MCP server for financial data: lessons learned

Originally published on Medium. I maintain bcb-br-mcp, an open-source MCP server that gives AI...

信息来源：DEV Community

教程 / 实战普通2026-09-27 04:27

A worktree per agent, so they never step on each other

By Andrew Pyle The first time I tried to run two coding agents at once on the same project, it went...

信息来源：DEV Community

教程 / 实战普通2026-09-27 04:22

在 LLM 盛行的时代，如何保持对编程的热情

一位 Haskell 程序员提出，用 LLM 做规划、调研等枯燥工作，而把写代码留给自己，以避免技能退化和代码库变成只有 coding agent 能维护的"LLM 荒原"。他建议让 LLM 把专家对话转成可执行 todo、用 markdown frontmatter 跟踪计划项，但关键决策必须由人来做，调研时自己也应并行搜索、至少大致掌握 agent 所知的一切。

信息来源：Hacker News 热门（buzzing.cc 中文翻译）

教程 / 实战普通2026-09-27 03:50

Astra Killed Claude, Claude Killed Gemini, Gemini Killed Google, Google Committed Suicide

Astra killed Claude. Claude killed Gemini. Gemini killed Google. Google committed suicide, and...

信息来源：DEV Community · Claude / Gemini

教程 / 实战普通2026-09-27 03:28

System One models in an agent loop: classify first, authorise in code

An agent needs a boundary between proposing an action and taking it. I would put that boundary in...

信息来源：DEV Community

教程 / 实战普通2026-09-27 03:28

Agent memory that tells you when it couldn't check

An on-call agent that answers "no prior incidents" might be telling you the archive is empty. It...

信息来源：DEV Community

教程 / 实战普通2026-09-27 02:45

I Built a Snowflake RAG Assistant That Actually Works in Production

Table of Contents The problem The architecture The Implementation Creating an...

信息来源：DEV Community

教程 / 实战普通2026-09-27 02:43

The check that passed while broken

Two checks in my build were green for months while testing nothing. One was switched off by a guard clause, the other by a missing space in a regex.

信息来源：DEV Community

教程 / 实战普通2026-09-27 02:40

Deny rules are not a boundary: building a policy hook for Claude Code and Codex

Coding agents such as Claude Code and Codex run with your shell and your credentials. Between October...

信息来源：DEV Community · Claude

教程 / 实战普通2026-09-27 02:05

Incident War Room: an incident-response tool where the approval gate and the AI can't be faked out

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2026-09-27 01:45

Mecropolis: Tracking Crop Stages from Past Weather

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2026-09-27 01:33

Not Everything Needs an AI Agent

Why adding autonomy to a software problem can make the architecture worse We have started calling...

信息来源：DEV Community

教程 / 实战普通2026-09-27 01:30

A security researcher told me to close my OAuth registration endpoint. I said no.

Last week I got an unsolicited security report about my QR code service. It was polite, it was...

信息来源：DEV Community

教程 / 实战普通2026-09-27 01:26

[AI in Action] Gemini Agentic Video: 4 Hidden Prerequisites and My LINE Bot Integration Process

Background I have a LINE Bot that I use every day, linebot-helper-python. If you send it...

信息来源：DEV Community · Gemini

教程 / 实战普通2026-09-27 01:25

[AI in Practice] Gemini 3.8 Flash TTS Launch: I built a "Learn Japanese with MVs" Web App and burned through my daily quota.

前情提要每次看到 Gemini 出新功能，我第一個念頭都是「能不能接進我的 LINE Bot」。 9/22 的 Gemini API release notes 寫著...

信息来源：DEV Community · Gemini

教程 / 实战普通2026-09-27 01:24

[AI in Practice] Deploying Song Lingo to Cloud Run: Making a Private Lyrics Website Just for Me

前情提要上一篇我用 Gemini 3.8 Flash TTS 做了 Song Lingo：貼 YouTube MV 網址，Gemini 轉錄歌詞、加上拼音翻譯文法，再由一位用...

信息来源：DEV Community · Gemini

教程 / 实战普通2026-09-27 01:23

Behind EcoPrompt: How I modeled the hidden physical cost of AI prompts (Zero Telemetry)

A deep dive into why I built EcoPrompt, the peer-reviewed research behind its metrics, and how to track LLM environmental impact locally in your browser.

信息来源：DEV Community

教程 / 实战普通2026-09-27 01:18

Writing code character by character

I want to talk a bit about where we came from in software and where we are now. Not the industry in...

信息来源：DEV Community

教程 / 实战普通2026-09-27 01:06

How Do You Know When AI Is Wrong? The Case for Evals.

Evaluation is not a new discipline invented for a new kind of system. It is the oldest discipline in...

信息来源：DEV Community

教程 / 实战普通2026-09-27 00:21

Self-Replicating Prompt Injections Turn Agent Context into an Open Relay

Most developers still treat prompt injection as a leakage problem. Someone types an adversarial...

信息来源：DEV Community

教程 / 实战普通2026-09-27 00:10

Agent Runtimes Have an Autoscaler, Not a Scheduler

Agent runtimes optimize for density and reject when full. That is not scheduling. Three failure modes at capacity, and who gets to call an agent...

信息来源：DEV Community

教程 / 实战普通2026-09-27 00:09

LoRA & DoRA: The Math, Memory, and Trade-offs

LoRA and DoRA explained: matrix shapes, the geometry of weight decomposition, exact memory math for a 27B model, and what the DoRA benchmarks support.

信息来源：DEV Community

教程 / 实战普通2026-09-27 00:09

I Built a Local RAG Pipeline with TypeScript, PostgreSQL and pgvector

Originally published on my personal...

信息来源：DEV Community

教程 / 实战普通2026-09-26 23:47

10 Real Airflow Incidents, 5 AI Models — Which One Handled Production Best?

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

信息来源：DEV Community

教程 / 实战普通2026-09-26 23:38

One operator, a fleet of agents

The honest constraint behind everything I build is that there is one of me. A portfolio of sites...

信息来源：DEV Community

教程 / 实战普通2026-09-26 23:17

I Built a Telegram AI Agent That Can Actually Use Tools O.O

I've been building Hexzie (HexTelegram) a Telegram-based AI assistant that goes beyond just sending...

信息来源：DEV Community

教程 / 实战普通2026-09-26 23:08

Claude Code said "Done." My tests said otherwise. Here's the 20-line fix.

A Claude Code Stop hook that runs your tests every time Claude tries to finish, and sends it back to work when they fail.

信息来源：DEV Community · Claude

教程 / 实战普通2026-09-26 23:03

Why Is Jev So Fast Compared to Traditional LLMs?

Most LLMs are built to generate. Jev is built to decide. That difference sounds small, but it...

信息来源：DEV Community

教程 / 实战普通2026-09-26 23:02

How to query Brazilian economic data with AI using MCP (Selic, IPCA, exchange rates, Focus)

Resumo em português: este tutorial mostra como dar a um agente de IA (Claude, Cursor, ChatGPT) acesso...

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

教程 / 实战普通2026-09-26 22:48

I Built a Local-AI Character Forge and a Browser Game Engine on a $100/y VPS (No GPU, No Excuses)

There's a special kind of stubbornness that leads someone to say "I will generate photorealistic 3D...

信息来源：DEV Community

教程 / 实战普通2026-09-26 22:48

AI Created a Leaked Credentials Flood: Here's How We're Draining It

TL;DR The exposure problem: AI-driven development pushed exposed credentials to 1.27...

信息来源：DEV Community

教程 / 实战普通2026-09-26 22:46

Your Company Does Not Need Another AI Chatbot. It Needs a Knowledge Layer.

Every company adopting AI eventually runs into the same problem. An employee opens ChatGPT, Claude,...

信息来源：DEV Community · ChatGPT / Claude