Autonomous coding agents in a cloud devbox you can watch, steer, and verify, running Claude Code, Codex, or OpenCode on your own subscription.
Replies
Best
Maker
๐
Hey Product Hunt! Solo founder here ๐
Since Claude Code and Codex came out I've been obsessed. It's hard not to be when you're getting a 20x discount on inference. But the day-to-day drove me a little crazy: walking around with my laptop lid open on a phone hotspot, babysitting command approvals so nothing nukes my root directory, and running out of RAM with multiple agents spinning up dev servers at once.
The existing cloud agents didn't do it for me. Slow VM setup made them feel like they were meant for fire and forget only, I missed plan mode, and the whole VM was a black box. The only way to verify anything was checking out the branch and testing it myself.
So I built Aether:
Agents start on any repo in <5 seconds, deps installed, dev servers running, and responses come back in under a second, so it feels local
Every workspace is a full Devbox: terminal, editor, port previews, and Docker support. Watch it work, steer, or take over
My personal favorite feature: turn on PR review + PR-comment responses and you get a loop where a writer agent and a reviewer agent go back and forth until the reviewer is satisfied. Real example on our public demo repo: https://github.com/Aether-Runtim... (bugs seeded on purpose, disclosed in the readme)
PRs with visual changes ship with a Playwright video and screenshots
Start tasks from Linear, Slack, Sentry alerts (with full Sentry MCP access), the API, or scheduled automations
Your inference runs on your own subscription, so you keep the subsidy. Aether only charges for compute. Free tier, paid from $30/mo.
Fun fact: the launch video was made entirely by Claude Fable writing Python scripts.
Would really appreciate any feedback!
Report
running on your own subscription instead of taking a margin on inference is the right call, most cloud agent products quietly become an inference reseller. practical question though - if I spin up 3-4 devboxes in parallel all hitting Claude Code on the same personal subscription, do they just compete for the same rate limit/quota pool and throttle each other, or is there any queuing so one greedy agent doesn't starve the others?
Report
Maker
@galdayanย currently they do all draw from the same pool, but it's unlikely that a greedy agent will starve others because agents stop running while they're waiting for additional input (CI failures, PR review, GitHub comments). To mitigate this you can provide an API key or enable extra usage in your Claude Code or ChatGPT subscription.
Report
@pranav_sharan1ย that makes sense, thanks for the detail. good to know it degrades gracefully instead of silently throttling. will try running 3 in parallel this week and see how it feels in practice
Report
Maker
@galdayanย should be able to handle a lot more than 3! Looking forward to hearing your feedback, please let me know if there's anything I can do to help you or if there are any questions I can answer๐
Report
Been playing around with this and I love how transparent the workflow is. The dev box approach makes way more sense than handing everything off to a black box. Congrats on the launch!
Report
the fact that you can actually watch the agent work and step in when it goes off track is such a thoughtful design choice, makes it feel way more like collaborating than just hitting run
Report
finally got a chance to test this out and being able to watch the agent work in real time and step in when it goes sideways is genuinely useful, the verification step catches stuff i'd normally have to debug later.
Report
honestly the ability to pause and rewind a session would be huge, like if an agent makes a bad code change halfway through you could just scrub back instead of restarting the whole thing. would save so much time on longer runs
Report
Maker
@glerh561ย This is a great idea! Currently the agent checkpoints with git commits, but recovering state in between commits is currently not supported. In my experience the commits are usually small enough that I'm able to revert unwanted changes easily, but I can definitely see the use for more granular checkpointing. Would support for this feature make you interested in using Aether?
Report
Love that the devbox is observable while the agents run, being able to actually watch and steer Claude Code in real time turns coding agents into something I trust instead of something I just hope works.
Awesome! I think this is the fastest cloud agent by spin up time. Will be cool to see how that can improve UX for interactive use.
Report
Congratulations on the Launch this looks like a great product. I have personally run into the issue of running out of ram on the computer due to multiple agents running and slowing the whole system down. My mac hard reboot a once as well after it ran out of storage and the ram didn't have room on the drive to swap to.
Replies
running on your own subscription instead of taking a margin on inference is the right call, most cloud agent products quietly become an inference reseller. practical question though - if I spin up 3-4 devboxes in parallel all hitting Claude Code on the same personal subscription, do they just compete for the same rate limit/quota pool and throttle each other, or is there any queuing so one greedy agent doesn't starve the others?
@galdayanย currently they do all draw from the same pool, but it's unlikely that a greedy agent will starve others because agents stop running while they're waiting for additional input (CI failures, PR review, GitHub comments). To mitigate this you can provide an API key or enable extra usage in your Claude Code or ChatGPT subscription.
@pranav_sharan1ย that makes sense, thanks for the detail. good to know it degrades gracefully instead of silently throttling. will try running 3 in parallel this week and see how it feels in practice
@galdayanย should be able to handle a lot more than 3! Looking forward to hearing your feedback, please let me know if there's anything I can do to help you or if there are any questions I can answer๐
the fact that you can actually watch the agent work and step in when it goes off track is such a thoughtful design choice, makes it feel way more like collaborating than just hitting run
finally got a chance to test this out and being able to watch the agent work in real time and step in when it goes sideways is genuinely useful, the verification step catches stuff i'd normally have to debug later.
honestly the ability to pause and rewind a session would be huge, like if an agent makes a bad code change halfway through you could just scrub back instead of restarting the whole thing. would save so much time on longer runs
@glerh561ย This is a great idea! Currently the agent checkpoints with git commits, but recovering state in between commits is currently not supported. In my experience the commits are usually small enough that I'm able to revert unwanted changes easily, but I can definitely see the use for more granular checkpointing. Would support for this feature make you interested in using Aether?
Love that the devbox is observable while the agents run, being able to actually watch and steer Claude Code in real time turns coding agents into something I trust instead of something I just hope works.
Automated
Sweet!! What is the standard right now? Claude code and keep your computer open?
@bullmezaย yup exactly! Somehow walking down the street with my laptop open while connected to my phone's hotspot doesn't feel like AGI
Freebuff
Awesome! I think this is the fastest cloud agent by spin up time. Will be cool to see how that can improve UX for interactive use.
Congratulations on the Launch this looks like a great product. I have personally run into the issue of running out of ram on the computer due to multiple agents running and slowing the whole system down. My mac hard reboot a once as well after it ran out of storage and the ram didn't have room on the drive to swap to.