Robot Learning Research Engineer (Manipulation Policies)

You’ll develop the learning systems that give our robots adaptability in the real world. From data pipelines to training and deployment of ML models , you’ll push the frontier of embodied intelligence by making robots learn directly from large-scale, real-world experience., * Design and implement ML algorithms for robot perception, control, and decision-making.

• Build and maintain data pipelines for large-scale collection, labeling, and training from robots in the field.
• Develop real-time inference systems that integrate seamlessly with the robotics stack.
• Collaborate with controls, perception, and embedded teams to deliver robust end-to-end autonomy.
• Optimize models for reliability, latency, and performance on hardware-constrained platforms.
• Push forward new approaches in robot learning - imitation learning, RL, foundation models for robotics., * Direct ownership of the learning systems powering humanoid robots.
• Early equity with meaningful upside in a venture-backed robotics company.
• The ability to ship ML models to live robots in weeks, not quarters.
• Exposure to the full robotics stack - hardware, controls, perception, embedded systems.
• A front-row seat in scaling a technically ambitious company from seed stage.

• Strong proficiency in Python and experience with ML frameworks (PyTorch, TensorFlow, JAX).
• Hands-on experience with robotic software stacks (ROS2, real-time control, perception pipelines).
• Background in reinforcement learning, imitation learning, or embodied AI .
• Experience deploying ML models to real-world systems, not just simulation.
• Strong debugging skills across hardware, software, and data.
• Bonus: Publications in robotics/ML conferences (RSS, ICRA, CoRL, NeurIPS), or experience scaling large data-driven systems.

Mundane is a venture-backed seed-stage robot learning startup founded by a team of Stanford researchers and builders. We’re deploying a massive fleet of humanoid robots to perform mundane tasks in commercial environments, collecting data to build the next generation of embodied intelligence. We’re a fast-paced, execution-driven team of engineers, roboticists, and dreamers.