I was hoping for a whole catalogue of providers - doesn't
amp have their own offering here "Orbs"?
So, the interesting part for me was:
"Who runs the fleet Anthropic, its own E2B, rented, third party"
Like that's the only option, a bespoke platform or a rented SaaS for agent sandboxes.
Disclaimer: I am tooting my own horn here - we built SlicerVM.com (Firecracker + Apple Virtualization framework) so that people don't have to pick between those two options.
Especially when you have your own 1500-7000 USD MacBook right under your fingers.
For remote access, (and this happened by accident one night in January) we built superterm.dev - initially as a tmux manager, then as a way to run and monitor agents. Our team sits with 6-24 long running tasks with dictation available, great mobile experience, and can combine both for secure agents + high amounts of feedback.
If anyone's curious - Superterm has a CE edition - free for unlimited personal use. Slicer has a free trial and then a personal edition is available after that.
E2B is fully open-source under Apache 2.0 and runs on GCP, AWS, Azure (preview) and locally on any Linux box that has /dev/kvm (Mac works via Colima, too).
yes i havent tried out the rest of them, i learned about the space mainly just by poking around these 2 agents that I use. It seems very fun though since theres lot of optimization problems at play. Ill check out your one too would be cool to see what apples virtualization framework is like!
Enjoyed the breakdown, honestly most of the things weren't really surprising.
What was surprising to me was that instinct was just "agent slapped on top of markdown files in git", especially after seeing people talk about it so much. I expected at least a database or a graph or something "smart", but alas, its just three markdowns in a trenchcoat.
well to be fair this was just what i could reach inside the sandbox by grepping around if they are doing something else behind the api i wouldnt be able to see it
in guessing the most important thing to get right for their target usecase is the computer use and from the tools i saw for that the computers they provision were requestable by the subagents so probably something cool going on there
I took a look inside the VMs that a couple of the agent platforms are using.
TLDR is they use firecracker with memory in git.
I'm a big fan of the memory in git part it would be cool to see how the git history looks after I use it for a longer period of time.
This is my first blog, I'm trying to document things publically as I learn, next I'll look more closely at the providers they are using under the hood (eg. e2b) so feel free to give advice on what to look for!
my thoughts is that Firecracker might be not the best runtime for that, as it’s very minimal on devices it wants to support (famously no PCI out of the box)
Woah so cool! A bit tangential but I think a lot of us on HN are open source fans, do you have advice on how to find an open source idea which has potential to grow into a large project like yours?
Jason Calacanis (famous venture capitalist, podcaster, early investor into Uber) said in the last all-in that there were humans doing stuff for you in the Instinct.
I also thought all the permissions I had to give it were scary, given how big of a single point of failure it'd be.
Can you share whether it's possible for your customer to have automated indians doing jobs as part of their implementations of the VM?
I deleted my Instinct after one day of productive work and I hope they finished my data.
claude -p "${curl $url}\nreply only with a substantive comment in 'statement. question?' form" | dd conv=lcase
as a service
i wonder how hard it would be to write a decent classifier for individual comments like this (in practice you can just look at the account history which would make it an easy problem)
I was hoping for a whole catalogue of providers - doesn't amp have their own offering here "Orbs"?
So, the interesting part for me was:
"Who runs the fleet Anthropic, its own E2B, rented, third party"
Like that's the only option, a bespoke platform or a rented SaaS for agent sandboxes.
Disclaimer: I am tooting my own horn here - we built SlicerVM.com (Firecracker + Apple Virtualization framework) so that people don't have to pick between those two options.
Especially when you have your own 1500-7000 USD MacBook right under your fingers.
For remote access, (and this happened by accident one night in January) we built superterm.dev - initially as a tmux manager, then as a way to run and monitor agents. Our team sits with 6-24 long running tasks with dictation available, great mobile experience, and can combine both for secure agents + high amounts of feedback.
If anyone's curious - Superterm has a CE edition - free for unlimited personal use. Slicer has a free trial and then a personal edition is available after that.
Amp’s Orbs run on E2B.
E2B is fully open-source under Apache 2.0 and runs on GCP, AWS, Azure (preview) and locally on any Linux box that has /dev/kvm (Mac works via Colima, too).
source: https://github.com/e2b-dev/infra
yes i havent tried out the rest of them, i learned about the space mainly just by poking around these 2 agents that I use. It seems very fun though since theres lot of optimization problems at play. Ill check out your one too would be cool to see what apples virtualization framework is like!
[flagged]
It's always, firecracker all the way down!
Enjoyed the breakdown, honestly most of the things weren't really surprising.
What was surprising to me was that instinct was just "agent slapped on top of markdown files in git", especially after seeing people talk about it so much. I expected at least a database or a graph or something "smart", but alas, its just three markdowns in a trenchcoat.
well to be fair this was just what i could reach inside the sandbox by grepping around if they are doing something else behind the api i wouldnt be able to see it
in guessing the most important thing to get right for their target usecase is the computer use and from the tools i saw for that the computers they provision were requestable by the subagents so probably something cool going on there
I took a look inside the VMs that a couple of the agent platforms are using. TLDR is they use firecracker with memory in git. I'm a big fan of the memory in git part it would be cool to see how the git history looks after I use it for a longer period of time. This is my first blog, I'm trying to document things publically as I learn, next I'll look more closely at the providers they are using under the hood (eg. e2b) so feel free to give advice on what to look for!
hey, i work at E2B, happy to answer any questions!
super cool product, any thoughts on adding native llm support to your sandboxes ?
my thoughts is that Firecracker might be not the best runtime for that, as it’s very minimal on devices it wants to support (famously no PCI out of the box)
you guys are awesome!
How did you find their internal services ?
accessed the vm through my terminal through a websocket: https://github.com/RohanAdwankar/ws-term
so basically you convince the agent to run a script which allows you to get a “shell” and then from there i just grepped around and figured it out
Do you have an invite to instinct? :)
yes i think everyone gets 10 did u want one?
Yes, how can I reach out to you to get one :) thanks
does dms via https://x.com/rohanadwankar work?
Thank you. I think invite got cleared up and I got one automatically.
Hi, if possible, could I get one too? You can contact me at esquire-zoom-jimmy@duck.com
[dead]
Hey everyone, I'm the CEO of the E2B - the VMs that the article talks about.
Happy to answer your questions!
Woah so cool! A bit tangential but I think a lot of us on HN are open source fans, do you have advice on how to find an open source idea which has potential to grow into a large project like yours?
I'd suggest look at your own problems. Tomas (my co-founder) and I built E2B back in 2023 because we needed a cloud computer for our agent back then.
What led you to use Firecracker over something like Kata containers?
We wanted a cloud computer for our agent that would be as close to a metal box as possible. No docker, no containers, etc.
Many of our users run containers and even k8s or k3s inside the sandbox. This is much easier to do when you have a VM and not container inside the VM
Jason Calacanis (famous venture capitalist, podcaster, early investor into Uber) said in the last all-in that there were humans doing stuff for you in the Instinct.
I also thought all the permissions I had to give it were scary, given how big of a single point of failure it'd be.
Can you share whether it's possible for your customer to have automated indians doing jobs as part of their implementations of the VM?
I deleted my Instinct after one day of productive work and I hope they finished my data.
this reminds me that i'm long overdue for a weekend spent studying firecracker, the userfaultfd trick [0], and whatever zeroboot does
[0]: https://codesandbox.io/blog/how-we-clone-a-running-vm-in-2-s...
Wondering how you determined the E2B sandbox is Firecracker and not some other micro VM?
[flagged]
[flagged]
[flagged]
[dead]
[dead]
[flagged]
i wonder how hard it would be to write a decent classifier for individual comments like this (in practice you can just look at the account history which would make it an easy problem)
Good spot, total bot account. Does HN have a way to report?
[flagged]