I’ve been obsessed with the idea of autonomous computer agents ever since Elon Musk tweeted about “computer control agents", I was like why am I still clicking through the same repetitive steps across machines, configs, and apps when we now have models powerful enough to just do it for us?
So I built llmhub.dev. It’s basically you can spin up on real machines. You tell them what to do, they figure out which LLM to use, and then they execute safely in a virtual machine.
Right now, every user can run up to 2 virtual machines with 5 CPU cores, 5GB RAM, and 20GB of storage each. You connect instantly with one click, no setup required. Files and apps stay right where you left them between sessions, and you can upload/download from your own machine anytime. Think of it as giving them a proper computer, not just a prompt box.
We’re still early, but I use this daily to clear out boring setup, testing, and admin flows. It’s surprisingly freeing to delegate this stuff to an agent and just check the results.
If you want to try it, shoot us an email and we’ll set you up with a free $50/month plan to play around. Would love feedback on what workflows we should add, where the safety model could be tighter, and what integrations would make this actually useful for your day to day.
AMA!
0 comments
—
Show HN: Autonomous Computer Using Agents at Scale Controlling Real VMs · HackerTrans
So I built llmhub.dev. It’s basically you can spin up on real machines. You tell them what to do, they figure out which LLM to use, and then they execute safely in a virtual machine.
Right now, every user can run up to 2 virtual machines with 5 CPU cores, 5GB RAM, and 20GB of storage each. You connect instantly with one click, no setup required. Files and apps stay right where you left them between sessions, and you can upload/download from your own machine anytime. Think of it as giving them a proper computer, not just a prompt box.
We’re still early, but I use this daily to clear out boring setup, testing, and admin flows. It’s surprisingly freeing to delegate this stuff to an agent and just check the results.
If you want to try it, shoot us an email and we’ll set you up with a free $50/month plan to play around. Would love feedback on what workflows we should add, where the safety model could be tighter, and what integrations would make this actually useful for your day to day.
AMA!