Robot Learning Engineer: Sim-to-Real Expert

WORLD LABS; San Francisco (CA); On-site; USD 250,000 - 350,000; Full time

Job summary

World Labs is seeking a Robot Learning Engineer/Scientist to advance state-of-the-art methods for training robot policies, with a focus on sim-to-real transfer and scalable training pipelines. You will collaborate with researchers, ML engineers, and system teams to translate policy stacks into production-ready implementations, enabling robust, fast, and reliable robot behavior.

This role sits at the intersection of robotics and machine learning, offering opportunities to shape deployment stacks

Qualifications

• 6+ years experience in manipulation, locomotion, or robot policy training.
• Strong robotics foundation and sim-real transfer expertise.
• Experience with robot policy designs (e.g., VLA, WAM, diffusion).
• Proficiency in Python and/or C++, with hands-on robotics systems.
• Experience with deep learning (PyTorch) and low-level robotic controllers.
• Ability to work in ambiguous, fast-moving environments and drive projects from concept through deployment.
• Strong ownership, rigor, and measurable improvement mindset.
• Collaborative, high-caliber team player focused on quality.

Responsibilities

• Design and implement modern robot learning systems for manipulation.
• Prototype and productionize robotic policies prioritizing speed and accuracy.
• Develop training pipelines for sim-to-real transfer and domain randomization.
• Collaborate with simulation and infra teams to close the sim-to-real gap.
• Build end-to-end training/evaluation workflows for robot policies.
• Optimize training speed, latency, and data efficiency for production scale.
• Diagnose failures in simulation and real rollouts; improve robustness.
• Contribute to research direction and mentoring across the team.

Skills

Robot learning

Python

C++

PyTorch

Robotics

Research to production

Tools

ROS

Gazebo

Linux

Job description

World Labs is seeking a Robot Learning Engineer/Scientist to advance state-of-the-art methods for training robot policies, with a focus on sim-to-real transfer and scalable training pipelines. You will collaborate with researchers, ML engineers, and system teams to translate policy stacks into production-ready implementations, enabling robust, fast, and reliable robot behavior.

This role sits at the intersection of robotics and machine learning, offering opportunities to shape deployment stacks