Research Engineer – Robot Learning (AI & Robotics)

About the role

World Labs is seeking a Research Engineer / Scientist to advance robot learning methods. The role focuses on developing end‑to‑end robot policies, bridging simulation and real‑world deployment, and scaling training and inference pipelines.

Key responsibilities

• Design and implement modern robot learning systems, including imitation learning and reinforcement learning for manipulation.
• Research, prototype, and productionize robotic policies with emphasis on speed, precision, and scalability.
• Develop and improve training pipelines for sim‑to‑real transfer, using domain randomization, system identification, and real‑sim alignment.
• Collaborate with simulation and infrastructure teams to minimize the sim‑to‑real gap and integrate learning methods into real‑robot deployment stacks.
• Build end‑to‑end training and evaluation workflows, from large‑scale data generation to scaled training and evaluation.
• Optimize policy performance across the stack, including training speed, inference latency, and data generation efficiency.
• Diagnose failure modes in simulation and real‑world rollouts and design solutions to improve robustness and generalization.
• Contribute to technical direction by proposing research ideas, mentoring teammates, and setting best practices for robot learning.

Required profile

• 6+ years of experience in robot manipulation, locomotion, or robot policy training.
• Strong foundation in robotics, neural network design, and sim‑real transfer.
• Deep experience with robot policy designs such as VLA, WAM, or diffusion methods.
• Proficiency in Python and/or C++ with hands‑on experience building research or production robotic systems.
• Experience with deep learning frameworks (e.g., PyTorch) and low‑level robotic controllers.

Required skills

• Python
• C++
• PyTorch
• Imitation learning
• Reinforcement learning
• Robot policy design (VLA, WAM, diffusion)
• Simulation‑to‑real transfer
• Domain randomization
• System identification
• Low‑level robotic controller development