Product Manager, Agent Harness (Robotics)

About Menlo

Menlo Research is an Applied R&D lab building Asimov, an open-source humanoid robot platform, and the full software stack that powers it. Our mission is to make humanoid labor economically viable, turning software into physical labor at scale. We build across the full stack: hardware architecture, locomotion, autonomy, simulation, and infrastructure. We move fast, ship to real robots, and open-source everything we can. If you want your work to matter beyond a paper or a demo, this is the place.

The Role

Coding agents changed how software gets built. We’re building the equivalent for robots. Menlo’s agent harness for robotics is an agent platform that researches engineering solutions, helps design manufacturable robots, trains them in simulation to perform physical tasks, and helps deploy them into the real world.

As Product Manager for the Agent Harness, you’ll own what this platform becomes. We’re looking for a deeply technical PM who has helped build Codex, Claude Code, Cursor, Devin, or a comparable coding agent, and who can rapidly carry that experience into robotics and physical AI. You’ll work directly with technical users, researchers, and engineers, prototype workflows yourself, and turn an ambitious vision into releases that actually ship.

What You’ll Do

• Own the agent harness end to end, taking it from an ambitious vision to focused releases with measurable acceptance criteria across research and engineering

• Define how the agent plans, uses tools, manages context and state, executes, verifies its work, and recovers during long-running engineering and robotics tasks

• Shape how users delegate complex physical-world work to the agent, inspect results, intervene, and recover from errors, and turn those needs into clear product behavior and a delightful developer experience

• Work directly with technical users to find the initial use cases where the harness creates real value, across engineering research, robot design, simulation training, and deployment

• Drive agent reliability through evaluations, user feedback, and telemetry, raising task completion, reducing human intervention, and managing latency and compute

• Prototype agent workflows, integrate tools, and investigate failures yourself through code and execution traces

• Get fluent in robotics fast, from VLAs and world-action models to simulation and manufacturability, working alongside our Brain, Hardware, and Platform specialists to make informed product calls

What We Look For

• You’ve built coding agents, having worked on Codex, Claude Code, Cursor, Devin, Lovable, or a comparable coding agent as a PM, engineer, or technical founder, with meaningful ownership of shipped capabilities

• You understand agent systems deeply and can explain the design decisions and tradeoffs behind planning, tool use, context and state management, execution environments, verification, and recovery during long-running tasks

• You’re hands-on: you navigate unfamiliar codebases, prototype agent workflows, integrate tools, and independently investigate failures through code and execution traces

• You’ve made agents measurably more reliable, using evaluations, user feedback, and telemetry to improve task completion, reduce human intervention, and manage latency and compute

• You have strong product judgment for autonomous workflows, understanding how users delegate complex work, inspect results, intervene, and recover from errors

• You learn unfamiliar technical domains fast through papers, code, experiments, and collaboration with specialists, and you’re ready to build the robotics knowledge needed to make informed product decisions

• You own outcomes, working directly with technical users and turning ambitious ideas into focused releases with measurable acceptance criteria

Nice to Have

• Familiarity with robotics and physical AI, such as inverse kinematics, motion planning, control, reinforcement learning, imitation learning, vision-language-action models (VLAs), or world-action models (WAMs)

• Experience with room scanning, 3D reconstruction, scene understanding, or converting real environments into simulation assets

• Experience with simulation and robot learning infrastructure, including training pipelines, reproducible experiments, skill composition, and validation on physical robots

• Familiarity with robot design and manufacturability, including CAD, component selection, bills of materials, assembly, sourcing, and cost constraints

Why Join Menlo

You’ll define what a coding agent becomes when its output is a working robot. The harness touches every part of how Asimov gets built: the research, the design, the training, and the deployment. The category doesn’t exist yet, the team ships to real hardware every week, and we build in the open. If you’ve helped build the agents that changed software and want to do it again for the physical world, come build it with us.

A Note on AI

You don't need deep AI expertise for every role, but we do expect everyone at Menlo to be intellectually curious, drawn to tinkering and discovery, and excited to use AI as a real collaborator in their work. For some roles, AI fluency is a core requirement. When that's the case, we'll say so explicitly in the qualifications. People who thrive here don't treat AI as a novelty. They use it to think better, and make their work easier for others to build on.

Equal Opportunity and Accommodations

We hire talented people from a wide range of backgrounds. If you're excited about a role but don't meet every bullet, we still encourage you to apply. Menlo Research is an equal opportunity employer and does not discriminate on the basis of any legally protected characteristic. Menlo provides reasonable accommodations during the application process. If you need one, please let your recruiter know.