AWS Lambda MicroVMs for isolated execution of user and AI-generated code
aws.amazon.com
2 threads
I’m working on an OSS version of something very similar to this, but the angle is more sovereignty and security over AI agents. Will post on HN when it’s ready.
I think that you must be using firecracker given that Amazon itself is using that in this.
What does that...mean?
- how does it compare to modal
- can it run gpu tasks?