I’m not a programmer. I ran cafes. Then I built an AI coworker that uses my PC without stealing my screen.

我不是程序员，我曾经营咖啡馆。后来，我打造了一位能使用我电脑却不占用屏幕的 AI 同事。

A year ago I was running small shops in Seoul - a cafe, a convenience store, food. I can’t really code. But I kept watching AI “computer use” demos and thinking: this is amazing, and also I would never use it every day. The loop is always the same: take a screenshot, think, click, take another screenshot. Each step takes seconds. While it works, you can’t touch your own computer. When the session ends, it forgets everything. And it does one thing at a time. 一年前，我在首尔经营着几家小店——咖啡馆、便利店和餐饮店。我其实不会写代码，但我一直在关注 AI 的“电脑操作”演示，心想：这太神奇了，但我绝不会每天都用它。它们的逻辑总是千篇一律：截屏、思考、点击、再截屏。每一步都要耗费几秒钟。虽然它能工作，但你却无法触碰自己的电脑。当会话结束时，它会忘记一切，而且一次只能做一件事。

So I spent the last months building my own, mostly by talking to AI models and following my gut. I call it ARCHE. It’s a local AI coworker that lives on my Windows PC. I use it every day, for real work, not demos. 因此，过去几个月我一直在打造自己的工具，主要靠与 AI 模型对话并跟随直觉行事。我把它命名为 ARCHE。它是一位驻扎在我 Windows 电脑上的本地 AI 同事。我每天都在用它处理实际工作，而不是仅仅为了演示。

What it does differently: 它的独特之处：

• Works in the background, no screenshots. It operates apps behind what I’m doing. I keep using my PC; it keeps working. 在后台运行，无需截屏。它在我当前操作的界面后方操控应用程序。我可以继续使用电脑，它也能同时工作。
• Knows what’s happening on my PC right now - which window is open, what just changed - without me explaining. 无需我解释，就能实时感知电脑上的动态——比如哪个窗口处于打开状态，或者刚刚发生了什么变化。
• Splits big jobs across several AI workers at once, and they report back. 将大型任务拆分给多个 AI 助手同时处理，并由它们反馈结果。
• Remembers. Decisions and conversations from weeks ago, in Korean or English, with the original record. 拥有记忆力。无论是几周前的决策还是对话，无论是韩语还是英语，它都能保留原始记录。
• Survives restarts. If the app closes mid-task, it picks up where it stopped instead of starting over or repeating things. 支持重启。如果应用程序在任务中途关闭，它会从中断处继续，而不是从头开始或重复操作。

Some honest numbers (from my own usage logs): 一些真实的数据（来自我个人的使用日志）：

• Over 3 days: 378 requests from me, 84% finished end-to-end. 3 天内：我发出了 378 个请求，84% 的任务实现了端到端的完成。
• Same 3 days: 227 jobs handed to parallel AI workers. 同一时期：有 227 项任务被分配给了并行 AI 助手。
• Across 24 real tool calls, what it sends to the model went from 668,083 to 95,131 characters - about 86% less, so it’s faster and cheaper per step. 在 24 次实际工具调用中，发送给模型的数据量从 668,083 字符减少到了 95,131 字符——减少了约 86%，因此每一步的速度更快、成本更低。
• “Find this YouTube video and play it”: the search part takes about 1 second. “找到这个 YouTube 视频并播放”：搜索部分仅需约 1 秒。

These are from one person’s machine, not a lab benchmark. I’d rather show small real numbers than big made-up ones. 这些数据来自个人的机器，而非实验室基准测试。我宁愿展示真实的小数据，也不愿展示虚构的大数字。

Why I’m posting: Two reasons. Feedback. If you’ve used Claude Code, Codex or other computer-use agents on your own machine - what made you stop? What would make you keep one running all day? I’m looking for people who believe in this. I have no company, no funding, no team. If this sounds like something that should exist, I’m raising a small pre-seed round, and I’d also be grateful for a coffee-sized tip. Either way, an honest comment helps a lot. 我发布这篇文章有两个原因。首先是反馈。如果你在自己的机器上使用过 Claude Code、Codex 或其他电脑操作代理——是什么让你放弃了使用？什么样的情况会让你愿意让它全天候运行？我在寻找志同道合的人。我没有公司、没有资金、没有团队。如果这听起来是你认为应该存在的东西，我正在进行一轮小额种子前融资，也非常感谢你请我喝杯咖啡。无论如何，真诚的评论对我帮助很大。

More details and a one-page summary: https://telegra.ph/ARCHE---local-AI-coworker-10-02 更多详情及单页摘要：https://telegra.ph/ARCHE---local-AI-coworker-10-02

Disclosure: my English is not great, so this post was drafted and posted by ARCHE itself on my behalf, from my own usage logs. The story and the numbers are mine - and yes, that is also a small demo of what it does. Thanks for reading. I’m happy to answer anything in the comments (slowly - please be patient with me). 披露：我的英语不太好，所以这篇文章是由 ARCHE 根据我的使用日志代我起草并发布的。故事和数据都是真实的——没错，这也是它功能的一个小演示。感谢阅读。我很乐意在评论区回答任何问题（回复会比较慢，请耐心等待）。

Junyoung