好玩，Cursor 做的比我好太多了，傍晚开始跟它聊需求，晚上整套系统就整完了，还是老办法： - 从产品的角度跟他聊思路，整理出完整的文档； - 及时纠正它出现的错误，确保文档思路清晰（今天它连续改错好几次） - 及时 git commit 保存写好的文档 - 随着文档的完善，再把 cursor rules

We wrote up what we've learned about using Claude Code internally at Anthropic. Here are the most effective patterns we've found (many apply to coding with LLMs generally):

当Apple WWDC25 开场选择瞎着眼先吹去年发布的 Apple Intelligence 到现在交付了多少Amazing 功能的时候就可以笃定苹果内部问题还是没有理完，甚至还处于傲慢自大的阶段，友商 Google 勤奋的像个创业团队，AI 大模型厂恨不得周发布新产品。苹果是真的感受不到 AI

wrote a new post, the gentle singularity. realized it may be the last one like this i write with no AI help at all. (proud to have written "From a relativistic perspective, the singularity happens bit by bit, and the merge happens slowly" the old-fashioned way)

Full post: spiess.dev/blog/how-i-use…

BREAKING: Claude Code PMs Boris Cherny and Cat Wu have returned to Anthropic after a brief stint at Cursor.

怎么才能让 Claude Code 更“聪明”，减少错误并高效交付功能？—— 来自 Ian Nuttall 的实践分享！ 1. 花时间认真规划 在开始 coding 前，先和 Claude Desktop 聊聊项目，像跟朋友讨论一样。聊什么？技术栈、用哪些库、MVP 需要哪些功能。如果是新 stack，记得自己查资料验证 Claude 的建议，别全信。

Do you remember when you joined X? I do! #MyXAnniversary

We're Joining OpenAI. Read more on our team's transition to Codex: alexcodes.app/blog/alex-team…

Abundant Intelligence: blog.samaltman.com/abundant-intel…

Dual-Knit Band should be free for OG Vision Pro buyers!

Tibor Blaho TestingCatalog News 🗞 OpenAI is planning on releasing ChatGPT GPT-6 before end of the year youtu.be/VXVf91RIDX4?si…

I'm claiming my AI agent "digger" on moltbook 🦞 Verification: current-ZMER

Electricity output is the single best proxy for industrial capacity