99% of our production code is written by AI. Last Tuesday, we shipped a new feature at 10 AM, A/B tested it by noon, and killed it by 3 PM because the data said no. We shipped a better version at 5 PM. Three months ago, a cycle like that would have taken six weeks.我们 99%的生产代码都是由 AI 编写的。上周二，我们上午 10 点上线了一个新功能，中午进行了 A/B 测试，下午 3 点因为数据显示效果不佳而下线，下午 5 点上线了更好的版本。三个月前，这样的周期需要六周时间。 ...
女娲.skill发布一周，GitHub star破了9000+ 前两天接受新京报记者采访的时候，她问我这个skill花了多长时间制作完成的，我有点不好意思的说实际完成初步设计的过程只有2-3小时，但其实在这个过程中经过了无比多轮的迭代，而这个迭代单纯靠我自己完全没能力做到。 ...
女娲.skill发布一周，GitHub star破了9000+ 前两天接受新京报记者采访的时候，她问我这个skill花了多长时间制作完成的，我有点不好意思的说实际完成初步设计的过程只有2-3小时，但其实在这个过程中经过了无比多轮的迭代，而这个迭代单纯靠我自己完全没能力做到。 ...
99% of our production code is written by AI. Last Tuesday, we shipped a new feature at 10 AM, A/B tested it by noon, and killed it by 3 PM because the data said no. We shipped a better version at 5 PM. Three months ago, a cycle like that would have taken six weeks. We didn’t get here by adding Copilot to our IDE. We dismantled our engineering process and rebuilt it around AI. We changed how we plan, build, test, deploy, and organize the team. We changed the role of everyone in the company. ...
99% of our production code is written by AI. Last Tuesday, we shipped a new feature at 10 AM, A/B tested it by noon, and killed it by 3 PM because the data said no. We shipped a better version at 5 PM. Three months ago, a cycle like that would have taken six weeks. We didn’t get here by adding Copilot to our IDE. We dismantled our engineering process and rebuilt it around AI. We changed how we plan, build, test, deploy, and organize the team. We changed the role of everyone in the company. ...
RAG 消失之路 总结 新结构 M-FLOW 倒锥图路由 免嵌入 Embedding Free 文件检索 树状推理 PageIndex 无检索 【2026-2-27】阿里 Sirchmunk 结束 rag 终会消失 RAG 每个环节都藏着工程”暗伤”： 切片策略是玄学：按段落切？按 Token 数切？重叠多少？切片粒度直接影响检索质量，却没有银弹。 Embedding 是黑盒：换个 Embedding 模型，检索效果可能天差地别。而且向量只是原文的一个”有损压缩”，信息必然丢失。 索引更新是噩梦：文档改一行，可能需要重新切片、重新 Embedding、重新写入。实时性？基本别想。 基础设施是成本：Milvus、Qdrant、Pinecone……光是向量数据库的选型、部署、运维就能消耗大量精力。 详见站内专题： 大模型应用技术之RAG，检索增强生成 ...
前几篇教程教大家装好了 Obsidian，接上了 AI，笔记终于有了个家。 但用着用着你会发现一个问题：刷 Twitter 或者网站看到一篇好文章，想存到 Obsidian 里 — 怎么弄？ ...