{"slug": "launch-hn-machine0-yc-s26-persistent-cpu-and-gpu-vms-from-the-cli", "title": "Launch HN: machine0 (YC S26) – Persistent CPU and GPU VMs from the CLI", "summary": "Machine0 (YC S26), founded by Barnaby, launched a CLI tool that provides persistent CPU and GPU virtual machines for AI agents, with pricing starting at $0.013/hr for 1 vCPU/1 GB up to 60 vCPU/240 GB and GPUs like H100s and H200s, featuring 99.99% VM-level uptime, per-minute billing, and MCP integration. The tool is designed for long-horizon agent workloads, enabling agents to self-manage fleets of VMs, with use cases including agent fleets, RL environments, and product infrastructure.", "body_md": "Hi HN! I’m Barnaby, founder of machine0 ([https://machine0.io](https://machine0.io)). I’m building a CLI for long horizon agent compute: `machine0 new mybox` gives your agent a persistent cloud VM, billed by the minute, from $0.013/hr up to 60 vCPU / 240 GB RAM and GPUs (H100s, H200s etc), with 99.99% VM level uptime. Agents self drive via CLI or MCP.\n\nDemo: [https://www.youtube.com/watch?v=gyllkZ0M04E](https://www.youtube.com/watch?v=gyllkZ0M04E)\n\nAgent workloads are moving from ephemeral to always-on. A coding agent working on a complex feature runs 6-8 hours. Agent orchestrated training & RL runs take days. OpenClaw & Hermes run 24/7. As you run more in parallel:\n\n- Resources: a few agents on a large codebase saturate RAM and CPU. Model training and RL needs GPUs you don't have.\n\n- Security: `--yolo` on your personal machine is one prompt injection away from exfiltrated credentials.\n\n- Availability: close your laptop and the agent dies mid-task.\n\n- Isolation: there's no clean line between you and the minimum your agent actually needs.\n\nmachine0 gives every agent its own computer. It's a CLI simple enough that both humans and agents use it without reading docs: `machine0 new mybox` creates an SSH-ready VM with a static IP and HTTPS endpoint. Always on (with 99.99% VM level uptime) until you switch it off.\n\n- Billed by the minute. 1 vCPU / 1 GB at $0.013/hr up to 60 vCPU / 240 GB, plus GPUs from RTX 4000 Ada to 8×H200.\n\n- Suspend, snapshot and resume. Making it easy to pause your work, and come back to it later. Or to make a golden master image to stamp out clones for a fleet.\n\n- Block storage. Persistent volumes (from 10 GB to 16 TB) that you can manage with intuitive grammar: `--yolo` and attach to your VMs.\n\n- Profiles. Bundles of credentials, MCP connections, prompts, and env vars, injected at VM creation. So each agent gets exactly the capabilities you choose, and nothing else.\n\n- Agents self-serve. Hand the CLI or MCP server to Claude, Codex, or OpenCode and it manages its own fleet: spin up a box for a build, snapshot it, tear it down.\n\n- Reproducible Builds. Using NixOS flakes or Ansible playbooks with Ubuntu.\n\nHow do people use it today?\n\n- Agent fleets. People run a pilot agent that scopes work and delegates it to sub-agents, each on its own VM: shape a project with the pilot, and the workers implement it and open PRs. One customer runs hundreds of machines at once, spun up and torn down from the CLI.\n\n- Model optimization & RL environments. ML teams use machine0 for agent-orchestrated RL environments and model optimization work. One customer runs RL environments on 60 vCPU machines that stay up for days at a time; another keeps a suspended H100 around and points an agent at it overnight to grind on inference-speed optimizations.\n\n- Product infrastructure. One customer builds their product on top of machine0 rather than using it themselves: every user session gets a fresh XL machine from a versioned image of their own agent runtime. They've shipped hundreds of versions of that image and launched thousands of machines, most alive for two minutes.\n\nWhat’s under the hood?\n\nEvery machine is a full KVM virtual machine, not a container or sandbox. You get the real GPU exposed to the guest with its actual driver, kernel-level access (load any module or driver you want), and no syscall-interception layer between you and the hardware. The stack itself is deliberately dull: TypeScript, Postgres, Redis. We weigh heavily towards security, reliability and performance making machine0 ideal for sustained compute intensive workloads. About me\n\nI've been building cloud infrastructure for about 15 years. I dropped out of a PhD at Imperial College London on cloud resource allocation, later spent six years as co-founder and CTO of Upflow (YC W20), owning DevOps, infra and security personally the whole way to 7-figures in ARR because it was too high-stakes to delegate. machine0 started as a tool for me, I’m my own first user :) Asks\n\nWould love you to try it out and give us your feedback (see below). Or if you’re a company looking for compute for software factories, model training or RL environments, feel free to reach out at barnaby@machine0.io\n\n``` bash\n  # install machine0 \n  $ curl -LsSf https://machine0.io/install.sh | sh\n\n  # create a machine and ssh in\n  $ machine0 new myvm\n  $ machine0 ssh myvm\n```\n\nComments URL: [https://news.ycombinator.com/item?id=49348136](https://news.ycombinator.com/item?id=49348136)\n\nPoints: 3\n\n# Comments: 0", "url": "https://wpnews.pro/news/launch-hn-machine0-yc-s26-persistent-cpu-and-gpu-vms-from-the-cli", "canonical_source": "https://machine0.io", "published_at": "2026-08-18 16:26:42+00:00", "updated_at": "2026-08-18 16:42:02.104442+00:00", "lang": "en", "topics": ["ai-infrastructure", "ai-tools", "ai-agents", "developer-tools"], "entities": ["machine0", "Barnaby", "YC S26", "OpenClaw", "Hermes", "Claude", "Codex", "OpenCode"], "alternates": {"html": "https://wpnews.pro/news/launch-hn-machine0-yc-s26-persistent-cpu-and-gpu-vms-from-the-cli", "markdown": "https://wpnews.pro/news/launch-hn-machine0-yc-s26-persistent-cpu-and-gpu-vms-from-the-cli.md", "text": "https://wpnews.pro/news/launch-hn-machine0-yc-s26-persistent-cpu-and-gpu-vms-from-the-cli.txt", "jsonld": "https://wpnews.pro/news/launch-hn-machine0-yc-s26-persistent-cpu-and-gpu-vms-from-the-cli.jsonld"}}