What is Menlo
Menlo is an Applied R&D lab enabling a humanoid robot labor force.
Our Mission
Menlo’s mission is to:
- Enable an agent developer to configure a humanoid robot to do economically valuable work in 5 minutes, without specialized robotics expertise.
- Bring the average cost of training a humanoid to under $10,000, per economically valuable task.
- Bring the annualized TCO of a humanoid robot to under $30,000/year.
Our Approach
-
Agent-native Robotics — We treat humanoid agents as deployable robots with permissions, safety constraints, and operational telemetry - leveraging the rapid iteration cycles of the agent ecosystem to drive the development of robotics.
-
5-minute Deployment loop — Menlo accelerates deployment cycles through an integrated stack (Training → Simulation → Deployment → Telemetry → Redeploy) that creates materially faster iteration than fragmented toolchains.
-
Cost-collapse levers — We focus on cost-collapse levers that make training (i.e. failure) cheap: simulate first, validate in digital twins, then deploy. Every real-world failure costs time, money, and hardware—simulation reduces that risk dramatically.
-
Open Hardware Ecosystem — We deliberately avoid vertical integration on manufacturing. Hardware is sourced from an open supply chain, so we can focus our investment where it matters most: the intelligence stack.
Why a Humanoid labor force?
Humanity has always been constrained by labor. Every civilization is limited by how many people can work, who can work, and what those people can physically do. This shapes what we build, what we afford, what we dream.
An unlimited humanoid labor force removes that constraint. When robots can do economically valuable physical work—reliably, safely, at scale—the bottleneck shifts from “can we find enough workers?” to “what should we build?” Labor stops being scarce and becomes infrastructure.
Humanoids specifically matter because we don’t need to retrofit the world for them. They use our tools, walk our floors, work in spaces designed for people.
The bottleneck today isn’t hardware—it’s deployment. Training, simulation, and reliable real-world operation remain unsolved. This is what Menlo is building.