What you can build on a computer per agent
Each guide says who it's for, why it pays, what it costs, and exactly how to do it: in the CLI, and in Python and TypeScript where the SDKs cover the step. Every feature it relies on shows whether it runs today.
All 14 guides
Pick the one closest to what you're building. Each ends with related guides.
Agent platforms, a sandbox per customer
Give every user or project of your product its own persistent computer, agent and preview URL, and bill them for exactly what they use.
See the stepsParallel agent fleets
Prepare one environment, fork it into many exact copies, and give each copy its own task. Keep the winner, delete the rest.
See the stepsA coding agent on your repository, with a spending cap
Hand an issue to Claude Code, Codex or OpenCode in a clean VM, follow it live, cap what it can spend, and get a pull request back.
See the stepsBrowser automation and scraping at scale
Run real Chrome with persistent profiles on a real desktop, record sessions for audit, and fork a logged-in browser into many.
See the stepsCI runners with a whole VM
Run tests that need Docker, systemd or KVM on a fresh VM per job, from GitHub Actions or any CI, billed by the second.
See the stepsLong-running autonomous agents
Let an agent work on a task for days, with its files, tools and browser intact, waking on a schedule or a webhook.
See the stepsUntrusted and agent-written code
Run code you didn't write behind its own kernel, with secrets kept out of the machine and a firewall the code can't remove.
See the stepsHybrid local and cloud agents
Keep your coding agent on your laptop and send heavy builds, Docker stacks and parallel tests to HareBox, so the laptop stays quiet.
See the stepsPreview environments and live demos
Give every branch, pull request or prospect a running app at a private HTTPS URL that wakes when someone opens it.
See the stepsEvals and benchmarks for agents
Run thousands of agent tasks from one template, each in its own VM, and know the cost of every run.
See the stepsMobile apps and emulators
Boot an Android emulator, Firecracker or QEMU inside the sandbox through nested KVM, for tests and agents that use apps.
See the stepsData and research agents
Give an analysis agent notebooks, databases and the open web, run it on a schedule, and get results pushed to you.
See the stepsWindows apps through Wine
Automate a legacy Windows tool on a Linux desktop, driven by an agent through computer use.
See the stepsTeaching, workshops and interviews
Give every student or candidate an identical machine from a template, with a budget, and watch their desktop when they ask for help.
See the steps
What these have in common
Most guides fall into four patterns. Each replaces something a team would otherwise build and run itself: VM fleets, networking, snapshots, desktops and credential isolation.
- Platforms resell a computer per user, metered per sandbox, with their own margin on top.
- Fleets turn one prepared machine into many workers, so a backlog is worked on in parallel.
- Automation replaces fragile scripts with agents that can use a real browser and desktop.
- Engineering teams move CI, previews and heavy builds off laptops and onto per-second VMs.