---
title: 'Claude Code Can Now Coordinate Parallel Threads?!'
source: 'https://youtube.com/watch?v=0Cw0JiHCGdY'
video_id: '0Cw0JiHCGdY'
date: 2026-11-17
duration_sec: 597
channel: 'Julian Goldie SEO'
---

# Claude Code Can Now Coordinate Parallel Threads?!

> Source: [Claude Code Can Now Coordinate Parallel Threads?!](https://youtube.com/watch?v=0Cw0JiHCGdY)

## Summary

Anthropic has released a major update to Claude Code's Projects feature, introducing a coordinator that manages parallel threads. This update shifts the burden of juggling multiple sessions from the user to Claude, allowing for more efficient task management. The video explains the new thread system, the /btw and /fork commands, and several smaller features, along with how to access the beta.

### Key Points

- **The Problem: Managing Multiple Tabs** [00:02] — Users often have many Claude Code tabs open and must manually track each one. The update aims to let Claude handle this coordination.
- **The Core Update: Coordinator in Projects** [00:42] — In a project, Claude coordinates tasks across parallel threads. Users talk to one chat, and Claude splits the work and runs pieces side-by-side, acting as a proactive manager.
- **What is a Thread?** [02:05] — A thread is one worker, a separate Claude Code session with its own context. It does one piece of work and reports back to the main chat. Most threads run in the cloud, so they continue even if you close your laptop.
- **Coordinator Visibility** [03:42] — The coordinator sees what threads report back, not every single step. Users can open a thread to read its full story and can type directly to steer it.
- **The /btw Command** [03:55] — In regular Claude Code, /btw asks a quick side question without history or tools. Inside projects, it sends the question to its own separate thread that can do real work, without interrupting the current task.
- **The /fork Command** [05:07] — In regular Claude Code, /fork copies the conversation into a new background session. Inside projects, it starts a new thread from the current one to try a different approach, while the original thread keeps running.
- **Four Smaller Features** [05:39] — 1) Reminders: Set reminders on threads. 2) Sharing: Share a thread as a read-only link. 3) Artifacts: Easier to find recent artifacts. 4) Project settings: Simpler local folder access.
- **Project Preferences** [06:53] — Users can tell Claude how to run the project, like 'run at most two threads at a time' or 'only post when something finishes.' These are preferences, not hard limits.
- **Access and Availability** [07:22] — It's a public beta rolling out in stages, currently for Pro and Max plans. Available at claude.ai/code, in the desktop app, and mobile app. Not in terminal, VS Code, or JetBrains yet. Users can join a waiting list.

### Conclusion

The update transforms Claude Code from a tool where the user is the project manager into one where Claude coordinates the work. This shift is significant for efficiency, though it's still in beta and users should expect changes.

## Transcript

How many Claude code tabs do you have open right now? Be honest. Three? Five? Ten? And you're the one keeping track of every single one. What if Claude did
that job for you? Today, there's a big update to projects in Claude code, and it changes who does the managing. I'll show you what a thread is, two I'll show you what a thread is, two commands called slash BTW and slash
fork, and four smaller features that are easy to miss. Then I'll tell you the one thing you have to do to get access. But first, let's look at what actually changed. So here's the update. The Claude Devs account from Anthropic just
said they made big updates to projects in Claude code. Here's the simple version. In a project, Claude coordinates your tasks across parallel threads. Parallel just means at the same time. So instead of you running five
chats and checking each one, you talk to one chat. Claude splits up the work and runs the pieces side by side. Why did Anthropic build it this way? As people get more comfortable with Claude code, they start running more and
more sessions at once. You start one task, then you open another thread for the next thing. That works, but then you're the one juggling everything. So the coordinator does the juggling. And here's what's new. The coordinator is
now much more proactive. The coordinator is the Claude in your main project chat. It answers your questions right there in the project chat, and it gives you updates on the status of your ongoing threads. So you can ask the project chat
things like which threads are running, which pull requests need review, what's blocked right now. And it answers in plain words. Think of a manager who checks in with you instead of you chasing everyone down. You can also just
paste in work as it comes up, like a bug report or a list of tasks. Claude starts or hands it to a thread that's already working in that area.
I keep saying the word thread though. So what is a thread? That's coming up in a second because everything else in this update builds on it. What's a thread? A thread is one worker. You tell the project what needs doing and Claude
starts a thread for each task. Each thread is its own separate Claude code session with its own context. It does one piece of work and then it reports back to the main chat. Most threads run in the cloud. That means they run on
Anthropic's computers, not yours. So they keep going even if you close your laptop. If a thread works on code, it uses its own branch and it can open a pull request when the work calls for it. You also get an overview pane next to
the chat. It shows which threads are still working, which ones are finished, and which one is waiting on you. So you can walk away, come back later, and know exactly where things stand. Quick pause before I show you the next part. We've
got a 30-day Claude code roadmap inside the AI Profit boardroom. It's a step-by-step plan so you learn the tool in order instead of guessing where to start. There are also four coaching calls every
week. So if you get stuck on Claude code, you can bring your question to a call and get unstuck. Think about what you just learned. Projects, coordinators, parallel threads, that's a lot of new ideas at once. A roadmap and
live help make new tools like this easier to pick up. Link in the description or go to aiprofitboardroom.com. Okay. One more thing. The coordinator sees
what the threads report back. It doesn't see every single step. So if you want the full story, open the thread and read what it did. You can even type to that thread directly and steer it yourself. Now here's where it gets fun. Two
commands. And inside projects, they work a little differently than you might expect. First one, /btw. It stands for by the way. In regular Claude code, /btw lets you ask a quick side question. It
history. And it can't use tools. It just answers from what's already in the conversation. That's handy when Claude is busy and you only need a quick answer. Inside projects, it works differently.
From what's been shared about this update, /btw sends your side question to its own separate thread. And that thread can do real work. Meanwhile, the thread you were in keeps going on its task. So a
side question isn't just a quick answer anymore. It can become real work off to the side, and you never have to stop anything. Picture it in general terms. You're watching a thread work and you think, oh, I should check something else
too. Instead of interrupting that thread, you send a /btw. thread, you send a /btw. Second command, /fork. In regular Claude code, /fork copies your current conversation into a new background
running. The copy starts with everything in the conversation so far, so you don't have to explain it again. Inside projects, /fork starts a new thread from the one you're in so you can try a different
approach. The original thread keeps running. Now you have two paths going at once. If both end up with pull requests, you can compare them and keep the one to remember it. BTW is for a side question. Fork is for
a side path. Same starting point, different direction. Now let's go through the other features. There are four. Number one, reminders. You can set a reminder on any thread, or you can just ask Claude to remind you. The
thread steps aside until the time you picked. Then it comes back and the coordinator brings it up in the project chat. Number two, sharing. You can now share one thread as read-only link with anyone who signed
into Claude. Read only means they can look, but they can't change anything. Early write-ups mention one limit. A thread that used your computer can't be shared. Number three, artifacts are easier to
find again. The things your threads make are simpler to get back to. Recent ones panel. Number four, project settings and local folder access are simpler now. Anthropic's posts don't spell out every
change, so I won't guess. But here's the idea. If a task needs something only your computer has, you can ask Claude to run that thread on your computer. That control. Here's a tip from the docs.
You can tell Claude how you want the project run. For example, run at most two threads at a time, or only post when something finishes or gets blocked, or something finishes or gets blocked, or give me a status update on every thread.
project memory. One heads up, they're preferences, not hard limits. Also, projects use your plan limits faster than a normal session. So it's worth knowing that before you send a big
batch of work. Now the big question, can you use it today? Maybe. It's a public beta and it's rolling out in stages. Right now it's for pro and max plans. Team and enterprise aren't included yet. You'll
find projects at claude.ai/code in the code tab of the desktop app and in the Claude mobile app. It's not in the terminal versus code or JetBrains yet. And here's the key part. If you don't see projects yet, you need to join
It's a short form at claude.com/form/projects. you get in earlier. So to wrap up, before this, running lots
of Claude sessions meant you were the project manager. You decided who works on what and you checked back on every session. Now, Claude does a lot of that. You send work to one chat. Claude starts the threads, tracks them, and tells you
what's going on. /btw handles side questions. /fork lets you try another questions. /fork lets you try another path. Reminders, share links, and easier artifacts rounded out. It's still a beta, so expect things to change. If
you're on an eligible plan, get on that waiting list today. And if you try it, tell me what you think in the comments. Here's the thing. Right now, there's a window where most business owners haven't figured this out yet, and that
window won't stay open forever. It's a good time to get ahead on learning Claude code. Inside the AI Profit Boardroom, you get daily tutorials so you can learn a little each day. You get four coaching
questions live. You get a prompt library so you're not starting from a blank page, and you get the member map, which is a map of what's inside so you can find the right thing fast. Link in the description or go to
aiprofitboardroom.com. If you want the full process, the SOPs, and over 100 AI use cases like this one, join the AI Success Lab. It's our free community. Links in the comments and description. You'll get all the video
notes from this episode plus access to 95,200 members. The AI Success Lab is a place to learn AI together and it's free to join. The video notes from this episode give you
the main points from today so you can follow along at your own pace and come back to them whenever you need. Join the AI Success Lab today and I'll Join the AI Success Lab today and I'll see you in the next video.
