One agent or a thousand. Same motion.coming
Running one agent is a script. Running a fleet of them — scheduled, triggered, watched — is infrastructure. Fleet will make the fiftieth agent as easy to start as the first, with the capacity ours to hold. It's part of Deployments, the top layer of the platform, and it's coming — the layers beneath it are further along, and one is live today.
Run a workforce, not a wall of boxes.
Today, going from one agent to many means orchestration you build and babysit. The fleet will make that our job — start them, schedule them, watch them from one board, and let the capacity underneath be someone else's problem to hold.
One or a thousand, same motion
Starting the thousandth agent will feel exactly like starting the first. Scale won't be a rewrite or a migration — it'll be a number, and the number will be ours to carry.
Schedules and triggers
Agents that wake on a cron, on an event, on a webhook — running when there's work and quiet when there isn't. You'll describe when; the fleet will handle the rest.
A board, not a battle
A fleet you glance at, not one you operate. What's running, what finished, what needs you — surfaced in one place, so the fleet is something you read rather than something you firefight.
Capacity is ours to hold
No pool to size, no autoscaler to tune, no 3 a.m. page when a hundred agents wake at once. Holding capacity for the whole fleet will be our job, on the key you already hold.
The top rung. Standing on the rest.
The fleet is one piece of Deployments — Layer 03, the top of the platform. It stands on the agent runtime, which stands on inference. The fleet is the furthest out; the ground it will run on is already being laid, and the bottom of it is live today.
Start on the inference layer today — your key will carry over.
Your agent doesn't change.
Everything underneath does.
Start free · no card · Starter from $5/mo