# Manus — Rebuilding Manus with Manus: What changed, what broke, and how we work now

- Company: Manus (manus.im)
- Announced: 2026-10-30
- Category: not stated
- Coverage: not counted
- Announcement: no
- Group: routine
- Source: https://manus.im/blog/rebuilding-manus-with-manus
- Record: https://forck.live/items/30865-rebuilding-manus-with-manus-what-changed-what-broke-and-how-we-work-now
- Subject: Manus

Manus co-founder Tao Zhang reports what changed inside the company after engineering, product and marketing teams began working alongside agents by default. The post is a report of an experiment in progress, not a definition of an "AI-native organization," and it includes what has not worked. Manus says adding AI to old ways of working sometimes added noise, and that things changed only when the loop closed. The company built an internal tool called Manus-mono, or Mono, a single entry point plus a manual written for the agent that describes what each repository does and how they fit together. Mono does not merge repositories; it also keeps a library of architecture decision records so every agent session starts with the reasoning behind past decisions. A push to the test environment used to take 5 to 10 minutes, and now the full Manus stack runs on a laptop with checks taking seconds. Mono can read logs, configuration and error reports in production with scoped permissions, and it runs real user flows with test accounts in a browser and a mobile simulator, while people still make approval decisions. Manus says one engineer commonly has 10 to 20 requests moving at once, and its seven-person mobile team maintains four apps across two products while keeping pace with web.

## Evidence

Verbatim from https://manus.im/blog/rebuilding-manus-with-manus:

> In late September, during the final push on Manus 2.0, I sent one of our engineers a few feature requests and bug reports on Slack. Later I went to check how he was getting on. His screen had a row of open agent sessions, some in Manus and some in Codex. I clicked into one. The prompt was my Slack message, pasted word for word.
> 
> A year ago, he would have read my message, worked out what I meant, and then written the code. Now an agent reads the request, finds the relevant code, and proposes a change. The engineer's job is to direct the work, check it, and own the result. Most of the time a request is running, not waiting on him. Today, it's common here for one engineer to have 10 to 20 requests moving at once.
> 
> I'm proud of how quickly our team has adapted. I'm also uneasy. Is this pace right? What haven't we noticed yet?
> 
> We build agents at Manus, so we hold ourselves to two rules. We use our own product for real work. And we keep looking for the places where AI stops working.
> 
> I'd like to document what happens inside a company when people and agents work side by side by default.

---

Record: https://forck.live/items/30865-rebuilding-manus-with-manus-what-changed-what-broke-and-how-we-work-now
Catalogue: https://forck.live/llms.txt
Current issue: https://forck.live/feed.md