Machine0
Machine0 provides on-demand cloud VMs with dedicated resources and per-minute billing, purpose-built for AI agents that operate via CLI or the Model Context Protocol (MCP). The platform offers up to 60 vCPUs, 240 GB RAM, and optional GPUs, with static public IPs and HTTPS endpoints at .mac0.io that eliminate NAT and tunnel complexity. Users can suspend VMs to freeze state and stop billing, or create golden images from any VM to spin up clones.
- Developer Tools
- Artificial Intelligence
- Software Only
Funding
Raised to date
$500KRaised to date based on public sources. This may differ from the amount the company actually raised and is based only on what is publicly available on the internet.
Founders
Founder details are not available yet.
Product
Problem
Developers running long-lived AI agents and compute-intensive workloads often face cloud infrastructure that is either too rigid for programmatic control or too expensive when billed by the hour. Managing ephemeral, multi-step agent tasks requires fine-grained lifecycle management—start, suspend, snapshot, and resume—without incurring costs for idle time.
Solution
Machine0 delivers on-demand cloud VMs with dedicated resources and per-minute billing, purpose-built for AI agents that need to operate via CLI or the Model Context Protocol (MCP). Every VM gets a static public IP and an HTTPS endpoint at `<vm>.mac0.io`, eliminating NAT and tunnel complexity. Users can freeze a VM's state with suspend, stop billing immediately, and restore later, or create golden images from any VM to spin up clones. Operations are scriptable via CLI commands with `--json` output, and a remote MCP server enables agents to drive the entire platform programmatically. Reproducible builds are supported through NixOS flakes for deterministic, drift-free environments, or Ubuntu images pre-installed with Docker, Node, Python, Claude Code, and Codex.
Target Audience
Primary customers are AI/ML engineers, DevOps teams, and independent developers running long-lived autonomous agents or batch compute jobs that require dedicated, scriptable infrastructure with granular cost control.
Features
- Up to 60 vCPUs, 240 GB RAM, and optional GPUs including H100, H200, L40S, MI300X, and RTX 4000/6000 Ada
- Per-minute billing with dedicated (non-shared) resources and static public IPs across five regions (US East, US West, UK, EU, Asia)
- Full lifecycle control: start, suspend, snapshot, and resume commands that freeze state and stop billing
- CLI-first design with `--json` output for every operation, enabling scripted automation
- Remote MCP server integration so AI agents can provision, manage, and tear down VMs programmatically
- Reproducible builds using NixOS flakes (deterministic, one-command rollbacks) or Ubuntu pre-baked with common AI tooling
- HTTPS endpoints at `<vm>.mac0.io` for each VM, with no NAT or tunneling required
- Open-source base images for NixOS and Ubuntu available on GitHub for transparency and customization