We're a small team with a big belief: that anyone should be able to give a robot a task the same way they'd ask a colleague. Come help us build that.
We are looking for talented engineers who want to work at the frontier of physical AI. We move fast, think deeply, and care about building things that actually work in the real world.
Build the agent layer that interprets natural language instructions and orchestrates robot actions across tasks and environments. You'll design the systems that turn "move this here" into a reliable, generalizable plan a robot can execute.
Develop and fine-tune VLA models that ground language commands in visual perception and physical robot control. You'll work at the intersection of foundation models and real-world robot deployment — where the hard problems live.
Design learning systems that let robots acquire new manipulation skills quickly — from demonstrations, language-described goals, or a combination of both. You'll build the pipelines that close the loop from human intent to robot behavior.
Create simulation environments and synthetic data pipelines to accelerate training and safe evaluation of robot policies at scale. You'll build the virtual worlds our robots learn in before they work in the real one.
Small team, large scope. You'll own significant parts of the system end-to-end and see your work running on real robots with real customers.
We care about staying at the frontier of the science and we care equally about the science working in production. Both are required.
We build for people, not around them. Every design decision is grounded in the question: does this make it easier for humans to work with robots?
Send us a note about yourself and your work. We read everything. No recruiters, just the team.
hello@lavoro.bot