Technology
HumanLayer
HumanLayer is the API and SDK for safe, production-grade AI: it embeds human-in-the-loop workflows for oversight, approvals, and feedback directly into autonomous agents.
HumanLayer delivers the critical infrastructure layer for deploying high-stakes AI agents safely. Our core product is an SDK that enables agents to pause and request human approval for sensitive actions, like a database administrator agent attempting to delete production SQL tables. Developers use the `@hl.require_approval()` decorator to enforce this oversight, ensuring reliability and accountability. The platform handles omnichannel routing, reaching the right team or individual via Slack, Email, or SMS for a prompt decision, and provides a full audit trail. Founded by Dexter Horthy, HumanLayer allows teams to scale powerful, autonomous AI without sacrificing control or building complex approval plumbing from scratch.
What builders pair with HumanLayer
Projects using both technologies. Select a pairing to see a project.
Pairing: Slack
Empowering AI Agents with Human Oversight: A Toolkit for Safe Tool Access
Pairing: Python
Empowering AI Agents with Human Oversight: A Toolkit for Safe Tool Access
Pairing: SQL
Empowering AI Agents with Human Oversight: A Toolkit for Safe Tool Access
Pairing: Email
Empowering AI Agents with Human Oversight: A Toolkit for Safe Tool Access
Pairing: FunctionLayer
Using HumanLayer to give agents safe access to high-stakes tools
Pairing: Python SDK
Adding human oversight to agent function calls with HumanLayer
Recent Talks & Demos
Showing 1-3 of 3