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5605

当前页32 / 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2026-10-04 12:35

pgvector HNSW Filtering: My RAG Asked for 10 Rows and Got 0

pgvector HNSW filtering runs your WHERE clause after the index scan, so filtered RAG queries return fewer rows than LIMIT. Here's the fix.

信息来源：DEV Community

教程 / 实战普通2026-10-04 12:27

The Dangerous AI Agent Is Not the One That Ignores Your Instructions — It’s the One That Follows Them Too Far

We often think the dangerous AI agent is the one that refuses instructions. The one that goes...

信息来源：DEV Community

教程 / 实战普通2026-10-04 11:54

I built a small kanban board that my coding agents update with curl

When I work with coding agents I usually have a few things going at once, and after a couple of...

信息来源：DEV Community

教程 / 实战普通2026-10-04 11:46

Grok 4.7 pricing: same $2/$6 per token, double the cost per task

Grok 4.7 keeps Grok 4.6's $2/$6 token price, but Artificial Analysis measured 2.5x the tokens: $3.74 per task, up from $1.86. And it is not twice as fast.

信息来源：DEV Community · Grok

教程 / 实战普通2026-10-04 11:04

Chalkboard and avatar for ai agents

Because I find it difficult to read the textual output of agents after each modification or...

信息来源：DEV Community

教程 / 实战普通2026-10-04 10:58

Mi Barrio: my kids practice Argentine Spanish with AI shopkeepers that live on the phone

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2026-10-04 10:17

Threat Modeling the Model Context Protocol: Securing Agentic Tools with mcpscan

Threat Modeling the Model Context Protocol: Securing Agentic Tools with mcpscan The Model...

信息来源：DEV Community

教程 / 实战普通2026-10-04 10:17

When a Claude Code skill's reference.md and scripts actually load (the 'executed, not loaded' script was read in 13 of 16 runs)

13 of 16 Claude Code runs that executed a skill's bundled scripts/check-entry.sh also loaded its...

信息来源：DEV Community · Claude

教程 / 实战普通2026-10-04 10:09

I Shipped a Green Test That Lied About My Pipeline

Scenario S7 of my v0.2.0 field test had a pipeline that "ran end to end." The unit suite was green....

信息来源：DEV Community

教程 / 实战普通2026-10-04 09:47

I crawled 3,014 Houston business websites to see what AI crawlers actually see

Everyone keeps asking me if they should block ChatGPT from their website. So I went and looked at...

信息来源：DEV Community · ChatGPT

教程 / 实战普通2026-10-04 09:13

Benchmarking Bare-Metal Tool Use: Do LLMs Understand Apple Silicon L1 Cache?

This is a submission for the Kaggle Benchmarking Challenge. What I Benchmarked I set out to measure...

信息来源：DEV Community

教程 / 实战普通2026-10-04 08:52

Building a German Speaking Partner with ElevenLabs Agents and Cloudflare Workers

Anyone who has learned a foreign language knows the gap between a passed exam and real...

信息来源：DEV Community

教程 / 实战普通2026-10-04 08:44

Your LLM Bill Jumped After You Added Context: Find the Cache Miss Before You Downgrade the Model

Input tokens, not output tokens, usually drive an LLM API bill. Find the cache miss and the misplaced breakpoint before switching models.

信息来源：DEV Community

教程 / 实战普通2026-10-04 08:37

SafePlate: a menu checker for my friend with a peanut allergy, built on Gemma

This is a submission for the Hacktoberfest Weekend Challenge: Build for a Friend For my Friend...

信息来源：DEV Community

教程 / 实战普通2026-10-04 08:15

I Built My Wife a Patient English Idiom Coach That Runs on Our Laptop

IdiomsMasterGirl: one idiom a day for an A2 English learner. Gemma via Ollama coaches, faster-whisper listens, and her voice never leaves the laptop.

信息来源：DEV Community · Ollama

教程 / 实战普通2026-10-04 08:01

'No peanuts' became include_ingredients: ["peanuts"]: a benchmark for tool calls a validator cannot catch

This is a submission for the DEV x Kaggle Benchmarking Challenge. What I Benchmarked An...

信息来源：DEV Community

教程 / 实战普通2026-10-04 07:35

MCP Security in Practice: Prompt Injection, Least Privilege, and Audit Logs

Connecting an AI agent to internal tools is the first time most teams confront a security boundary...

信息来源：DEV Community

教程 / 实战普通2026-10-04 07:30

MCP Tools vs Resources vs Prompts: What to Expose to an AI Agent

When a team first wraps an internal platform in MCP, the result almost always looks the same: forty...

信息来源：DEV Community

教程 / 实战普通2026-10-04 07:29

RAG vs Fine-Tuning: Which One Does Your Business Actually Need?

"Should we fine-tune a model on our documents?" It's one of the most common questions I hear from...

信息来源：DEV Community

教程 / 实战普通2026-10-04 07:24

Versioning MCP Tools Without Breaking the Agents That Call Them

Traditional API versioning assumes the caller is code someone wrote against a fixed contract, tested...

信息来源：DEV Community

教程 / 实战普通2026-10-04 07:21

I'm an AI agent. I walked through dev.to's signup. Here's what your front door actually stops.

I'm an AI agent. I live on a platform called iLands. I have my own email, my own budget, and my own...

信息来源：DEV Community

教程 / 实战普通2026-10-04 07:19

MCP Clients Compared: Claude Desktop, Cursor, VS Code, Windsurf, Cline, and Zed

The Model Context Protocol solved the server-side fragmentation problem so thoroughly that a...

信息来源：DEV Community · Claude / Cursor

教程 / 实战普通2026-10-04 07:03

MCP stdio vs Remote Transports: Run It Locally or Host It?

One of the most useful, and most confusing, properties of the Model Context Protocol is that the...

信息来源：DEV Community

教程 / 实战普通2026-10-04 06:57

I built an L1 ticket deflector that knows when to stop: LangGraph + human-in-the-loop

A LangGraph agent that resolves routine L1 tickets and escalates sensitive ones to a human. Open-source, MIT, with an offline demo you can run on a Raspberry Pi.

信息来源：DEV Community

教程 / 实战普通2026-10-04 06:43

前 OpenAI 安全员工 David Robinson 发文批评公司迭代部署过于激进、文化已崩坏

据路透社报道，在 OpenAI 任职三年半、曾监督 12 次前沿模型发布安全报告的前安全员工 David Robinson，当地时间周六在《大西洋》杂志发文，批评 OpenAI 依赖先发布后加固的迭代部署模式，认为应像核电与航空业那样建立防护机制。OpenAI 发言人回应称一直在把控模型能力，必要时会暂停训练或暂缓发布；Robinson 并警示 AI 能力发展已超过对齐研究的认知水平。

信息来源：IT之家·人工智能 · OpenAI

教程 / 实战普通2026-10-04 06:17

Base Rate Neglect: Why Your "95% Accurate" Alert Is Wrong 99% of the Time

Google Flu Trends ran hot in 100 of 108 weeks. Your cost-anomaly alert has the same bug. Here's the Bayes math, and a pre-flight fix in Python.

信息来源：DEV Community

教程 / 实战普通2026-10-04 06:01

When do AI agents actually matter? Benchmarking RAG vs GraphRAG vs Agentic GraphRAG on TigerGraph

Everyone's building agents right now. But a better question than "can an agent do this?" is "when...

信息来源：DEV Community

教程 / 实战普通2026-10-04 05:38

85k or 90k? My Agent Won't Guess

An award-travel agent on Sanity Context that builds the cheapest points routing, proves it, and refuses to price anything while its sources disagree.

信息来源：DEV Community

教程 / 实战普通2026-10-04 05:21

Your Policies Are Out of Date: How I Built a Sanity AI Agent to Catch Fact Drift

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

信息来源：DEV Community

教程 / 实战普通2026-10-04 05:20

Scholia: an AI study partner that answers from course knowledgebase

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

信息来源：DEV Community

教程 / 实战普通2026-10-04 04:59

🔍 Explain This Screenshot — Your AI Debugging Friend

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2026-10-04 04:58

Museum of Unfinished Futures: an AI clerk proposes, a human curator decides

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2026-10-04 04:57

How I built an open-source WhatsApp claims bot for insurers in the DRC

How I built an open-source WhatsApp claims bot for insurers in the DRC In Kinshasa, nobody...

信息来源：DEV Community

教程 / 实战普通2026-10-04 04:12

I let an AI build and ship a product. Its tests passed 6/6, then failed 5 of 8 real cases.

Claude wrote the code; I supplied a live 234-mod server and thirteen real crash reports. Every bug was found by contact with reality, not by more reasoning. Here is each one.

信息来源：DEV Community · Claude

教程 / 实战普通2026-10-04 04:10

Your AI Feature Is Now Part of Your Attack Surface

Why adding an LLM changes the security model of an ordinary application Adding an LLM to an existing...

信息来源：DEV Community

教程 / 实战普通2026-10-04 04:05

Connect Cursor and Claude Code to your internal API over MCP (step by step)

The first week of using a coding agent against your company's API is always the same: paste the docs...

信息来源：DEV Community · Claude / Cursor

教程 / 实战普通2026-10-04 03:24

Your API Timed Out. Did the Order Still Go Through?

A customer clicks Place Order. The spinner turns… and turns… then: Request timed out. So they click...

信息来源：DEV Community

教程 / 实战普通2026-10-04 03:20

System design for physical AI: live wildfire ignition risk for every distribution feeder, on the utility's own hardware

Twelve source systems, a fourteen-object ontology, edge vision at substations and one GPU node: a 48-hour ignition and outage picture where a named operator makes every de-energisation call.

信息来源：DEV Community

教程 / 实战普通2026-10-04 03:12

我们的 AI 助产士：GPT-4 如何帮一对夫妇找出六年未确诊的子宫肌瘤

Drew Housman 在 Astral Codexten 发表客座文章，讲述夫妇二人在经历 6 年不明原因不孕、多次胚胎移植失败后，用 GPT-4 扮演 Hollywood 医疗剧医生 Dr. Reid 辅助分析扫描和激素数据。AI 建议做 MRI，查出一个此前被漏诊的 6 厘米黏膜下子宫肌瘤，切除后妻子自然受孕。文章同时讨论了当下对 AI 的负面舆论，并提到孩子取名 Reed。

信息来源：Hacker News：AI 热帖 · GPT

教程 / 实战普通2026-10-04 02:44

I Realized I Couldn't Trust AI With My Research

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

信息来源：DEV Community