#AI编程
9632 related articles
Cursor Cloud Agent Now Supports Self-Hosting: Connect to Internal Services and Dedicated Hardware
Cursor now supports self-hosted cloud Agents on your own infrastructure, with auto-scaling machine pools for access to internal services and dedicated hardware.
Cursor + Origin + Vercel: A Complete AI-Powered Web Development Workflow
Cursor, Origin, and Vercel connect AI code generation, code hosting, and one-click deployment into a complete web app workflow. Here's what it means for developers.
Let AI Coding Agents Review PRs Without Granting Unrestricted Merge Access
How can AI coding agents review PRs without getting unrestricted merge access? Explore GitHub permission controls and the Airlock governance mechanism to prevent post-approval code tampering.
Google Accused of Copying Open-Source Code: 228 Out of 229 Files Identical, Engineer Credits Removed
A startup founder accused Google's open-source tool Artemis of copying 228 out of 229 files from their mobile-use project and removing engineer credits, sparking debate about open-source ethics.
Hands-On with Gemini 3.8 Live: Testing Google's Speech-to-Speech Model
Google launches Gemini 3.8 Live and Extended Thinking speech-to-speech models. We test the real-time voice experience and break down the zero-dependency WebSocket + Web Audio API implementation.
Claude Sonnet 4.5 Comes to Cursor: The New Coding Champion with 73.4% on CursorBench
Claude Sonnet 4.5 joins Cursor and scores 73.4% on CursorBench 3.2 in max effort mode — the highest ever. Its self-verification ability marks a leap toward autonomous coding agents.
Meta Muse Spark 1.3 Lands in Cursor: A New Option for AI-Assisted Coding
Meta's Muse Spark 1.3 is now available in the Cursor AI code editor. Learn what this integration means for developers and the AI coding tools ecosystem.
Gemini 2.5 Flash Lands on Cursor: Lightweight Model Accelerates AI Coding
Gemini 2.5 Flash is now live in Cursor. Learn how this lightweight, low-latency model enhances AI coding workflows and what it means for multi-model development.
DSH-SUBAGENT-UI Plugin: The Ultimate Sub-Agent Manager for DeepSeek Harness
DSH-SUBAGENT-UI is a DeepSeek Harness browser plugin offering sub-agent overview, search, local categorization, and completion snapshots — install with one command.
Cayu: An Open-Source Python Framework for Building Long-Horizon Domain-Specific Agents
Cayu is an open-source Python framework for building domain-specific, long-horizon AI agents. It lets developers assemble a harness of tools, knowledge, and business rules, backed by an integrated durable runtime handling sessions, state, recovery, approvals, and observability.
Meta Launches WhatsApp Business MCP Server, Letting AI Agents Handle Tedious Setup Tasks
Meta's new WhatsApp Business MCP server lets developers use AI agents like Claude, Cursor, Codex, and ChatGPT to automate API setup, message templates, testing, and troubleshooting.
Claude Code v2.1.273 Update: Permission Vulnerability Fixes and Enhanced Remote Control
Claude Code v2.1.273 update covers permission security fixes, remote session forking, MCP reconnection, improved error messages, and Slack & code review enhancements.
Perplexity's In-House CobbleDB: 2 Engineers + Hundreds of AI Agents Built Search Infrastructure in 2 Months
Perplexity reveals CobbleDB, an in-house key-value database built by just 2 engineers and hundreds of AI agents in two months, signaling a new era of AI-driven software development.
AI Agent Development: A 4-Stage Learning Roadmap from Beginner to Enterprise-Level Practice
A complete AI Agent learning roadmap from zero to enterprise-level: covering ReAct, multi-agent collaboration, Prompt tuning, RAG, MCP, and real-world projects.
A New DeepSeek Harness Experiment: Agent Supervising Agent for Self-Evolution
A developer built an "Agent supervising Agent" self-evolution system using DeepSeek Harness, with a ledger mechanism enabling near-unattended overnight software iteration.
From Solo to Team Player: The Journey of an AI/LLM Self-Learner
A CSE student with LLM, RAG, AI Agent, MCP, and LangChain skills seeks mentorship and team experience. Exploring how AI self-learners can transition from solo to collaborative engineering.
How DeepSeek Cut AI Inference Bills by 150x: A Four-Step KV Cache Compression Breakdown
A deep technical breakdown of how DeepSeek's new model slashes AI inference costs to 1/150th — from 890-byte KV cache entries to encoder-decoder splits, 3D compression, and moving n-gram tables off the GPU.
Same Model, Three AI Agents: DeepSeek Harness vs. ZCode vs. Hermes — Which One Wins?
Same GLM Flash model, same prompt — three AI Agent frameworks compared: DeepSeek Harness, ZCode, and Hermes. Results show framework choice matters more than you think.
DeepSeek Harness in Practice: Building a Low-Cost AI Coding Powerhouse
Learn how to transform DeepSeek's open-source harness using Claude Code, Bright Data scraping, and vision models to build an AI coding workflow costing just half a cent per task.
Overseas Developer Tests: DeepSeek Already Rivals Opus — Stop Waiting for the Next Model
An overseas developer finds DeepSeek V4 Pro rivals Opus 4.8 at a fraction of the cost. Learn how DeepSeek + BrightData compares to Claude Code for building SaaS.