Zach Lloyd

We did build a lot with the developer in mind, like some of the hero use cases that we have built for things like Ben said earlier, like jobs to clean up stale feature flags, jobs to update your documentation when you do releases, that kind of stuff.

And it's interesting to see

even though we kind of built it for developers, how much these automations are clicking with general knowledge workers.

So I think we'll invest in making it even easier, but it is pretty easy.

Like Ben said, you just kind of need to like,

put your faith in the agent to get you set up.

Like you just make it really tangible.

If you want to get set up, you can just do it directly from Warp.

You just type slash create environment and that kind of gets you going.

And then in Warp, you can just like, you can open a whole new tab that just fires off something in what we call cloud mode, which under the hood is running an OS agent.

So it's pretty easy, but I think we can make it even easier to be honest.

It does make sense.

We have a whole effort around this.

And actually, if you're on our preview build, I don't know if you are, Michael, you can...

We released an early version of this that only works locally, where you can explicitly invoke it by typing slash orchestrate and provide a task and provide some instructions on how to divide a task up amongst multiple agents.

or if you just use our normal like slash plan planning facility, we will now on the preview bill, try to decompose complicated tasks that can be done by multiple agents.

And it's very early for us in doing this, but what you actually see is you sort of have a pattern of one agent that is like the orchestrator, and then you get, you know, and other agents that are doing tasks.

And you can actually, because we can control the whole UI of Warp,

You don't have to use like Tmux or anything like that to do this.

It will just automatically create like one pane per agent and you can watch the agents work.