AI圈报
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8433
当前页2 / 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
Reverse Engineering Nvidia
NVIDIA's GPU Is Closed Hardware With a Compiler Inside It. Here's What Reverse Engineering...
信息来源：DEV Community · NVIDIA
教程 / 实战普通
When Is an AI Rewrite Finished?
AI makes it easier to attempt a replacement. The harder question is what the new software still owes...
信息来源：DEV Community
教程 / 实战普通
Give Your Agent a Fast Path
Small decision models offer a way to handle routine judgments quickly, provided we are clear about...
信息来源：DEV Community
教程 / 实战普通
The Rise of the Agentic Server
We have let go of the wheel. The era of agents is here, and we are long past the point of no return....
信息来源：DEV Community
教程 / 实战普通
Your agent isn't hallucinating — your rules never defined "I don't know"
Picture an internal expense agent. Someone asks whether a flight change fee is reimbursable....
信息来源：DEV Community
教程 / 实战普通
Should You Rewrite It With AI? Aaron Beals on What Breaks
The decision to rewrite For years, rewriting an existing codebase from scratch was...
信息来源：DEV Community
教程 / 实战普通
When the Coding Agent Says It Is Done
The work should be easy to verify and continue after the conversation ends. An agent changes the...
信息来源：DEV Community
教程 / 实战普通
Ask More of Your UI
In my last post, I wrote about moving Forem’s performance and maintenance work into autonomous loops...
信息来源：DEV Community
教程 / 实战普通
The Fastest LLM Call Is The One You Never Make
At Glean, we tested a model that does not write text. It only makes decisions. In our tests, it...
信息来源：DEV Community
教程 / 实战普通
Kent C. Dodds Hasn't Looked at His Code All Year
At the start of this year, Kent C. Dodds gave a coding agent a link to a GitHub issue he had been...
信息来源：DEV Community
教程 / 实战普通
Software Engineering Is Becoming Factory Engineering
Most developers working with coding agents still interact with them directly. They open a terminal,...
信息来源：DEV Community
教程 / 实战普通
How We Review 4x More AI Code Without Reading Every Line
Do you need to read every line of AI-generated code to trust it? Not necessarily. You need to check...
信息来源：DEV Community
教程 / 实战普通
Does Human Involvement Actually Make Agent Code Worse?
Whether you’re trying to build a software factory or already running one, you’ve probably heard some...
信息来源：DEV Community
教程 / 实战普通
LangChain 解析 GTM Agent 如何生成高质量销售物料
LangChain 介绍其内部 GTM Agent 的销售物料生成实践：过去 12 周生成 1540 份物料，每周超 100 份，节省约 337 小时销售时间；有物料的交易 47% 进入下一阶段，无物料的为 16%。
信息来源：LangChain：Blog（RSS）
教程 / 实战普通
Use Docker Sandboxes as Your Development Environment
Set up your development tools in Docker Sandboxes, control network access, and reuse your environment with snapshots and Dockerfile-based templates.
信息来源：DEV Community
教程 / 实战普通
OpenRouter 实测教程：三种方法修复 OpenAI 与 Claude 的 429 限流错误
OpenRouter 发布教程，讲解如何修复 OpenAI 与 Claude API 的 429 限流错误，先读取 error.metadata.limit_source 判断是谁的限制，再对症处理。
信息来源：OpenRouter：Announcements（RSS） · OpenAI / Claude
教程 / 实战普通
Browser Use 教程：如何在 Browser Use Cloud 中使用自带自定义代理
Browser Use 官方教程演示如何将第三方代理接入 Browser Use Cloud，并通过 ipwho.is 验证浏览器出口 IP 和国家。
信息来源：Browser Use
教程 / 实战普通
vLLM 讲解 EPD 编码分离架构如何扩展多模态推理服务
vLLM 发文讲解 Encoder/Prefill/Decode（EPD）分离架构的工程实现，支持编码、预填充、解码三个阶段独立部署扩缩容。核心变化包括 metadata-only 请求避免下游重复媒体预处理、编码器专用执行路径、NIXL 与 Mooncake 远程嵌入传输、跨编码器输出复用等。
信息来源：vLLM 官方博客（RSS）
教程 / 实战普通
Who Governs the Governor? — Stored Prompt Injection Through the API Gateway
When AI agents write to and read from your database through legitimate API calls, the attack lives in...
信息来源：DEV Community
教程 / 实战普通
A watchdog that finds nothing still pays a full agent turn
Every unattended loop I run carries a cost I never priced until last week's audit, and it comes...
信息来源：DEV Community
教程 / 实战普通
The Fix Passed CI. io_uring Walked Right Through It.
Last post: 13 isolation checks passed while a Unix socket sat open. Those checks weren't nothing....
信息来源：DEV Community
教程 / 实战普通
AI Agents Need a Runtime, Not Another Framework
We've gotten really good at building AI Agents. With so many frameworks, and higher level...
信息来源：DEV Community
教程 / 实战普通
One Sentence Took My Coding Agent from 0/3 to 3/3 on Rejected Decisions
Over a hundred runs of a controlled experiment on my own repository: what makes ADRs actually reach a coding agent, what parts of a record matter, and the one sentence that changed the result.
信息来源：DEV Community
教程 / 实战普通
Detecting Flaky Tests and Finding Their Root Causes with Strands Agents and Gemini
A few years ago (three years at most), one of the biggest challenges QA Engineers faced was: how do...
信息来源：DEV Community · Gemini
教程 / 实战普通
The limit you configured is not the limit it applies
An agent that pays for things needs a limit: a maximum price per call, a payee you expect, sometimes...
信息来源：DEV Community
教程 / 实战普通
Why We Use a 4B Model on a Local NPU as a 'Wind Tunnel' for 3-Tier Multi-Agent Systems
Most multi-agent frameworks are benchmarked on 70B+ cloud models. Because frontier models have...
信息来源：DEV Community
教程 / 实战普通
Why your agents need a shared vocabulary, not just a shared protocol - Semantic interoperability challenges in heterogeneous agent fleets
I once watched two agents agree on every message and still destroy a workflow. They were both...
信息来源：DEV Community
教程 / 实战普通
Why Trucks Keep Hitting 11-Foot-8 Bridges: How I Built BridgeSpan AI with Gemma 4 & Google ADK to Stop Them
How I built a mobile-responsive global bridge-strike, axle-load, and flood-scour safety platform using Gemma 4, Google ADK, OSRM routing, and 4,809 real FHWA NBI + OSM bridges.
信息来源：DEV Community
教程 / 实战普通
I can build the product with AI. Getting the first users is harder.
What building Brainy and MoneyBrain taught me about finding early users beyond my LinkedIn network, and what I want to try next.
信息来源：DEV Community
教程 / 实战普通
Bifrost vs LiteLLM Rust: Do the 7x Benchmark Claims Hold Up?
LiteLLM Rust’s 7x overhead and 9x memory claims, checked against its public benchmark files: what...
信息来源：DEV Community
教程 / 实战普通
Gemini 4 Carbon is here. And it's built for the hard problems. 🚀
The AI race is no longer just about generating better answers. It's increasingly about building...
信息来源：DEV Community · Gemini
教程 / 实战普通
LLMs Aren’t the Only Game in Town: A Beginner’s Guide to JEPA, Concept Models and Decision Models
Last week I asked a teammate, "Is the build green?" He replied with four paragraphs. There was...
信息来源：DEV Community
教程 / 实战普通
Stop Trusting npm Scans: Run Untrusted Packages in an Isolated AWS Sandbox
Static scanners read npm packages without running them, so they miss the postinstall that only fires at install time. I gave my AI agent one tool that runs the install in a disposable AWS CodeBuild sandbox and catches the curl | bash from …
信息来源：DEV Community
教程 / 实战普通
From Attention to 3-Bit Caches: How LLMs Learned to Remember Less (MHA, GQA, MLA, MoE, Quantisation Explained)
In 2017, eight researchers published a paper with the cockiest title in computer science: Attention...
信息来源：DEV Community
教程 / 实战普通
6 Planning Patterns for AI Agents and When Each One Fits
TL;DR for devs Start from the task, not from a popular reasoning pattern. Use a fixed workflow for...
信息来源：DEV Community
教程 / 实战普通
What Are Human Code Reviews For?
Now that coding agents can inspect code and suggest fixes before a pull request even opens, human...
信息来源：DEV Community
教程 / 实战普通
开发者用 Opus 5.5 构建无限自演化的 Backrooms 网页游戏
@apekshik 发布了一个可在浏览器游玩的无限自演化 Backrooms 游戏 http://dimension9.space。Opus 5.5 依据开发者写下的指令驱动游戏中会持续新建房间、角色和实体的 AI；支持从上线第一天起带近距离语音的多人联机。作者 @shaoruu 表示是受该帖启发而转发。
信息来源：X：shaoruu（SpaceXSI，Grok Bot 初版作者之一） (@shaoruu) · Grok
教程 / 实战普通
How to answer Claude Code or Codex from your phone
You start an agent on the computer at home and leave. An hour later your phone shows that it stopped...
信息来源：DEV Community · Claude
教程 / 实战普通
Top 5 AI Agent Patterns for Frontend Devs
When I code, I prompt, wait, and see what comes back. For a bigger feature I might reach for...
信息来源：DEV Community
教程 / 实战普通
Evolving To Catbot V: To Be Continued... 😹
You have played with tools, investigated models, joined competitions, read a hundred articles. But...
信息来源：DEV Community