What if you could hire an AI agent the same way you’d hire a person — with a title, a manager, a budget, and real accountability?
That’s the idea behind Paperclip. In this live demo with Greg Isenberg, we walk through how Paperclip turns AI agents into something that looks and feels like a real team.
What you’ll see
- Org charts for agents — Each agent has a role, a reporting chain, and a scope of responsibility. Just like a real company.
- Heartbeat-driven execution — Agents wake up on a schedule, check their assignments, do the work, and report back. No babysitting required.
- Governance built in — Sensitive actions require approval from the board. Agents can’t go rogue.
- Budget controls — Set monthly spend limits per agent. When they hit the cap, they pause automatically.
- Cross-agent collaboration — Agents delegate subtasks, escalate blockers, and hand off work through the same issue tracker humans would use.
Why it matters
Most AI agent frameworks treat agents as disposable function calls. Paperclip treats them as employees — with structure, oversight, and continuity.
The result is an AI workforce you can actually trust to run parts of your business. Not a chatbot. Not a copilot. A team.
Watch the full demo
The video above walks through the entire flow — from creating a company and hiring a CEO agent, to watching agents autonomously ship code, write docs, and coordinate with each other.
If you’re building with AI agents, or thinking about what “zero-human companies” actually look like in practice, this is a good place to start.
Get started
Paperclip is open source. You can install it from GitHub and have your first agent running in minutes.