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5611
当前页47 / 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
RepoMind: A Self-Evolving Code Review Agent That Remembers How Your Team Builds Software
RepoMind: A Self-Evolving Code Review Agent That Remembers How Your Team Builds Software Code review...
信息来源:DEV Community
教程 / 实战普通
Context Compression for Coding Agents Compresses the Wrong Side of the Prompt
Most engineering teams working on long-context agents hit the same billing wall around turn twenty. A...
信息来源:DEV Community
教程 / 实战普通
The Overnight Speed Loop: Asking AI Agents to Make Code 7x Faster
Here is a sentence that would have sounded insane eighteen months ago: a developer pointed a coding...
信息来源:DEV Community
教程 / 实战普通
I built a PDF reader where every note remembers its page
Every time I read a long PDF (a paper, a spec, a book chapter) ended up with the same mess: a list of...
信息来源:DEV Community
教程 / 实战普通
Building a Restaurant Finder With $0 in API Costs (OpenStreetMap, Open-Meteo, No Google Maps)
VibeEats is a small food-discovery app: pick a mood ("cozy," "date night," "quick bite"), or search...
信息来源:DEV Community
教程 / 实战普通
Three things AI agents did on the web, and what they mean for people who build agents
Three real cases show how AI agents can cause problems even when they work as designed — and six lessons for anyone building agents for the web.
信息来源:DEV Community
教程 / 实战普通
SatQuery Al: Building a Conversational System for Satellite-Image Analysis
SatQuery AI: Building a Conversational System for Satellite-Image Analysis I built SatQuery AI around...
信息来源:DEV Community
教程 / 实战普通
RAG always needs a dedicated vector database — challenged
Hot take: you don't need a vector database for RAG. Most teams reach for a purpose-built vector...
信息来源:DEV Community
教程 / 实战普通
How to Build an AI-Ready Customer Profile
Connecting an AI application to a CRM or a customer data warehouse is relatively easy. The harder...
信息来源:DEV Community
教程 / 实战普通
AI Coding Agent Security Flaws: Claude Code, Gemini, Codex
28 advisories for Claude Code in 13 months, a CVSS 10 for Gemini CLI: the flaws live in the harness, not the model. The 5 patterns and how to defend.
信息来源:DEV Community · Claude / Gemini
教程 / 实战普通
I Gave My Proposal Agent Hindsight of Every Lost Bid
Aura Memory takes an RFP, works out what the client is asking for, writes an 11-section proposal, and...
信息来源:DEV Community
教程 / 实战普通
The Verification Gap: We Automated Code Generation and Forgot to Scale Review
James Coombs is a design engineer who writes the code-review skills that AI agents run on pull...
信息来源:DEV Community
教程 / 实战普通
Your AI Policy Doesn't Run in Production. Your Gateway Does.
Why governance for LLM apps and agents is an infrastructure problem, and what to actually...
信息来源:DEV Community
教程 / 实战普通
Stop Paying for AI APIs: The Blueprint for a 100% Private, Local AI Stack
` The AI ecosystem is shifting rapidly from cloud‑only models to hybrid and fully local deployments....
信息来源:DEV Community
教程 / 实战普通
How I Killed 63 Flaky Tests in One Week With Claude Code: 5 Lessons
TL;DR I had 63 flaky tests in a Node.js monorepo that made every CI run a coin flip. Over...
信息来源:DEV Community · Claude
教程 / 实战普通
用NVIDIA Nemotron 3 Ultra搭建论文研究智能体
用Tavily拉取arXiv近7天论文,由部署在Nebius Token Factory上的NVIDIA Nemotron 3 Ultra读取并排序,约10秒即在终端得到本周智能体记忆方向Top 5论文。
信息来源:X:Elvis Saravia (@omarsar0, DAIR.AI) · NVIDIA
教程 / 实战普通
Why a Coding Agent Needs an MCP for Allure TestOps When a Human Can Just Open the Browser
Developers on our team switched to working through coding agents fairly quickly. In my experience,...
信息来源:DEV Community
教程 / 实战普通
Creem.io for Pakistani Developers: A Practical Integration Guide (2026)
READ HERE ...
信息来源:DEV Community
教程 / 实战普通
How LLMs Are Creating New Security Risks in CI/CD Pipelines
Traditional automation was deterministic and relatively predictable. LLMs work differently. They...
信息来源:DEV Community
教程 / 实战普通
I Replaced a Gate That Accepted Everyone With a Gate That Accepted No One. My Tests Couldn't Tell the Difference.
Run this in a terminal, then run it again under script(1): python3 -c 'import sys;...
信息来源:DEV Community
教程 / 实战普通
Is regex enough? I tested span-01 on mixed-language text
Have you ever exported a script for text-to-speech and found, later, that a single English word had...
信息来源:DEV Community
教程 / 实战普通
Three rebuilds: what building a trading platform with AI agents actually took
I started building Atlas on July 31. It's now on its third version, and the one running today finally...
信息来源:DEV Community
教程 / 实战普通
I asked an LLM to fix misspelled names. It renamed the product instead.
My AI clipper corrects proper nouns in transcripts. On one video it swapped an unknown model name for...
信息来源:DEV Community
教程 / 实战普通
One hotfix, three repos, four surprises: why I wrote a Git Flow doctor
We still run classic Git Flow at work: main, develop, and short-lived release/* and hotfix/*...
信息来源:DEV Community
教程 / 实战普通
A Technical Deep Dive into Enterprise AI Agents and Conversational Automation
Artificial intelligence is moving beyond simple chatbots and question-answering systems toward AI...
信息来源:DEV Community
教程 / 实战普通
I built a music-looping tool for TTRPGs and game prototypes with Python and Web Audio
The boss finishes their speech. Your players reach for their dice. This is where the music should...
信息来源:DEV Community
教程 / 实战普通
I ran claude plugin eval on 89 skills. The trigger rate fell from 100% to 55.6%.
The problem with skills is that failure is silent My coffee e-commerce project has 89...
信息来源:DEV Community · Claude
教程 / 实战普通
Traffic from ChatGPT now brings a quarter of our sign-ups. How we track it, and what we changed
In April, 3.7% of new members on my website came from ChatGPT. In September so far it is 24.5%, and last week it was one in three. Here is the share week by week, why we couldn't see it at first, and every change we made for AI search, in …
信息来源:DEV Community · ChatGPT
教程 / 实战普通
The Best Evidence for a Fact-Checking Process Is the Errors It Catches
AI assistance made my technical writing faster and my errors faster too. The pipeline that resulted — claim ledgers, publish-day drift guards, and a separate instance that checks the writer — caught three of its own mistakes before readers…
信息来源:DEV Community
教程 / 实战普通
I cut six streaming SDKs at every point. Nothing I found was new.
The last thing I wrote here was about two bugs my own test suite could not see. Both came from the...
信息来源:DEV Community
教程 / 实战普通
My On-Call Agent Remembered the Fix That Took Down Checkout
02:06 UTC checkout-api started timing out. I pasted the alert into a capable LLM and asked what to...
信息来源:DEV Community
教程 / 实战普通
Rapid Prototyping a Web Audio Saxophone Synth with Gemini and MIDI
How we spent an hour using Gemini and the Web Audio/MIDI APIs to rapidly build, test, and record a responsive wind controller synth using a real instrument.
信息来源:DEV Community · Gemini
教程 / 实战普通
Build a Contract Obligations Review App with ToolJet MCP
Introduction Here is how to build a contract obligations review app with ToolJet MCP for...
信息来源:DEV Community
教程 / 实战普通
Designing Hindsight Retain and Recall for SRE Incidents
An incident-response system can retrieve the right information and still fail to learn...
信息来源:DEV Community
教程 / 实战普通
I stripped metadata from AI images to see if YouTube would still flag them
I've been curious about this for a while. Every AI image generator I've tried writes metadata into...
信息来源:DEV Community
教程 / 实战普通
Agent history is unsigned and writable by anyone
Darktrace showed that coding agents trust their local conversation history without verifying it. A writable file is enough to give the model orders. Here is what to watch.
信息来源:DEV Community
教程 / 实战普通
LongCat 2.5 Preview vs Space Bunny Alpha: My Experience Building Real Projects
Over the past few days, I've been comparing LongCat 2.5 Preview and Space Bunny Alpha across actual...
信息来源:DEV Community
教程 / 实战普通
When AI Reviewers Become a CI Problem
Fifteen AI review bots on one repository is not fifteen times the insight — it is a governance problem. How one open-source project keeps machine reviewers advisory, deterministic gates blocking, and PR code out of the trust boundary.
信息来源:DEV Community
教程 / 实战普通
We Let Vibe Coders Merge to Production. We Kept the Core.
How non-engineers build AI tools on an internal MCP platform and self-merge them to production — the self-merge gate, clean-context AI code review, and the one zone that still needs an engineer.
信息来源:DEV Community
教程 / 实战普通
Your AI Agent Changed One Repo. Did It Check What Breaks Outside It?
Hello Devs 👋 One thing I have noticed with AI coding agents is that they are very good at working...
信息来源:DEV Community