Tag
AI coding
AI coding covers the use of LLMs, agents, and editor integrations across the dev loop: completion, refactoring, test generation, code review, and task execution. The real issue is not raw speed, but context handling, harness design, permissions, and verification in existing codebases.
25 articles
Vibe coding lets you ship a tiny app fast
A practical breakdown of vibe coding, with a copy-ready prompt workflow for building and fixing a small app in AI tools.
What Vibe Coding Means for Developers
Vibe coding uses plain-language prompts and AI agents to turn ideas into code faster, with humans still reviewing every step.
Anthropic Ships Opus 4.8 With Better Coding
Anthropic says Opus 4.8 improves coding and that new safety work could open mythos-level models to more users.
Cognition raises $1B at $25B pre-money valuation
Cognition raised more than $1 billion at a $25 billion pre-money valuation as Devin gains enterprise traction and revenue grows fast.
How GitHub’s outages stalled Microsoft’s AI coding lead
I break down why GitHub’s outages and Azure migration pain weakened Microsoft’s AI coding edge, and give you a copy-ready playbook.
Claude Sonnet 4.8: not released as of May 25
Claude Sonnet 4.8 is still unannounced. Anthropic’s current Sonnet is 4.6, and the 4.8 rumor rests on one leaked source string.
Google AI Studio brings Android app building to web
Google AI Studio can now generate native Android apps in minutes, then hand them off to Android Studio or Play Console.
Why AI Job Cuts Are Dumb
AI productivity should expand teams and output, not justify layoffs.
IBM’s vibe coding guide turns prompts into code
I break down IBM’s vibe coding guide into a practical workflow, its limits, and a copy-ready template for AI-assisted coding.
GitHub Copilot app enters desktop agent preview
Microsoft opened a desktop preview of GitHub Copilot for macOS, Windows, and Linux, with Agent Merge and paid-plan access.
Claude Code 生产级工作流完全指南
Anthropic 限制中国大陆直连后，开发者更常用国产 Claude 兼容接口来跑 Claude Code。
Why Anthropic’s SpaceX deal proves AI coding needs more compute, not …
Anthropic’s SpaceX deal shows AI coding progress now depends on massive compute, not product theater.
Cursor CEO Michael Truell’s rapid rise
Michael Truell turned Cursor into a $60 billion SpaceX deal after starting coding at 11 and leaving MIT for startups.
How Windsurf Flow Keeps Context Alive
Windsurf Flow updates AI context as you work. Here’s how RAG, Cascade, Memories, and rules shape every suggestion.
Cursor 3 Workspace: Multi-Agent Workflow in One Place
Cursor 3 moves agent work into one workspace, with parallel sessions, cloud handoff, and a new path from commit to PR.
AI coding is fast. Trust is the bottleneck
AI coding tools are shipping faster than humans can review them, pushing enterprises toward code governance, testing, and trust layers.
Claude Mythos Raises the Bar for AI Coding
Anthropic’s Claude Mythos pushes ahead in coding, reasoning, and security, while its cost and dual-use risks keep adoption selective.
Cursor 3 Adds Parallel Agents and Design Mode
Cursor 3 adds parallel agents, Design Mode, and self-hosted cloud agents, with pricing and enterprise controls now more explicit.
Claude Mythos vs Opus 4.6: How Big Is the Jump?
Leaked benchmarks suggest Claude Mythos could beat Opus 4.6 by a wide margin in coding, reasoning, and security tasks.
Cursor CEO warns vibe coding builds shaky software
Cursor CEO Michael Truell says skipping code review with AI makes software brittle, while Cursor keeps engineers in the loop.
Google’s Agent Smith Is Quietly Changing Coding
Google’s internal Agent Smith automates coding tasks, works in the background, and is already so popular access had to be limited.
Claude Code 里的 Harness Engineering 思路
Claude Code 把 MCP、Skills、Hooks 和 Subagents 直接暴露给用户，展示了 Anthropic 如何把 Harness Engineering 做进产品。
Cognition's Devin and the AI coding race
Cognition grew from 40 staff to 200+ after buying Windsurf, as Devin pushes harder into enterprise coding and AI software work.
Cursor Adds Self-Hosted Agents and Real-Time RL
Cursor shipped self-hosted cloud agents and real-time RL for Composer, with checkpoints updated as often as every five hours.
Cursor's Meteoric Rise Faces Industry Pressures
Cursor, a $30 billion AI startup, faces intense competition from Anthropic and OpenAI, challenging its rapid growth in the AI coding market.