probably: saves money vs a fleet of consistently running instances

> From a billing perspective, AWS does not charge for the EC2 instance itself when stopped, as there's no physical hardware being reserved; a stopped instance is just the configuration that will be used when the instance is started next. Note that you do pay for the root EBS volume though, as it's still consuming storage.

https://depot.dev/blog/faster-ec2-boot-time

Their product seems to be offering "buildkit as a service" and I'd guess from their perspective the safest isolation boundary is at the VM level. Unknown why they don't boot up a bigger .metal instance and do their own virtualization but I'm sure there are fine reasons

(You wouldn't)

Patting themselves on the back for 'fixing' a self-created problem... EC2 is the wrong abstraction for this use case imo