How we built forkable vm's in opencomputer.dev 00:11 Architecture 00:35 Zooming in to a single worker 01:13 Forkable VM's 02:43 Demo 04:10 Results 04:24 Typescript sdk

Do you believe that hallucinations and agents breakout can be solved by fixing up the models?

Timeout bound coding agent sandboxes make you babysit snapshots and preview URLs to keep an app "alive". In opencomputer.dev, when idle, the box hibernates ($0 compute, URL intact). Any request coming in auto wakes it. Hibernation snapshots the VM state (memory disk) automatically and stops the VM, so there are no compute costs while hibernated, you only pay for the stored snapshot. Folks building lovable's for {x} love this! docs.opencomputer.dev/sandbo…

With the abundance of open weights providers out there I really wanted to be able to have an API to compare the same model across providers as well as the ability to have all that information be consumed easily by LLM. I've created ai-pricing.fyi/ for exactly that reason. Any feedback is welcome!

We built four malicious skills to test whether skill scanners actually work. Three took less than an hour to conceive and implement. ClawHub, Cisco, and Vercel's skills.sh marked them as safe. 🧵

1/ After founding a gaming company and spending more than a decade in venture, one thing has become very clear to me: I love early-stage investing. Today, I’m excited to share that I’ve started @entropyVC.

@motatoeshq cooked with this. I've been hooked to clawputer over the last few days opencomputer.dev/clawputer uses: - opencomputer.dev as the computer - @pipedream for integrations - @browser_use for browser use (lol) - linqapp.com for imessage infra

Firecracker doesn't support live migration. It has snapshot/restore ie. pause the VM, save state, restore later. But it is not really the same thing. True live migration means serializing a running VM (CPU, memory, devices) and moving it to another host mid execution. QEMU has done this for 10 years w iterative pre/post-copy dirty page tracking. Firecracker was built for Lambda. Long running agents need both Lambda ergonomics WITH EC2 semantics. QEMU ftw.

The “spin it up, tear it down” sandbox model is very 2024. Agents don’t run for 200ms any more. They run for hours. Sometimes days.

Agentic AI’s blind spot is not just what agents can read. It’s what they can do after they read it: spawn processes, write files, call APIs, reach the network, and use credentials. That control point is the execution layer. That’s why we’re building AgentSH: agentsh.org

Want to be a 1-person code factory? @timoreilly and I chatted all about how it works oreilly.com/radar/ryan-carso…

same here. will bring nice things too.

Agents today are processes that live for hours. The substrate should match! Today, agents pause, resume, branch, retry, sit idle waiting for a human, then wake up and keep going. You can’t really run that on infra designed for stateless functions.

buckle up. next 2-3 weeks at opencomputer.dev might be the most fun stretch we’ve had as a company. so much shipping. you’re going to love it.

"my agent OOM'd on a rust cargo build" came up in basically every early discovery call we had for opencomputer. the nail in the coffin was a pickleball session in dogpatch with @tokengobbler where he detailed this out to the literal T haha. so we built our elasticity feature. the agent now hits 169.254.169.254 from inside the VM and scales its own RAM and CPU mid-task without a restart. agents are getting ambitious, so we think the substrate should, well, keep up? (set min=max and it's disabled, you can opt out if you want fixed sizing.)

No single winner. But the winners will nail the same things every infra primitive eventually demands: 1. Reliable af 2. Goated DX, devs should genuinely love using it 3. Opinionated (we’ve never really had an infra primitive that was everything for everyone) 4. Tasteful. code is cheap today. authenticity is the moat, and it’s brutally hard to fake.