Robotics Learning Engineer: RL/IL for Autonomous Navigation

ROZOR; Hamilton; On-site; CAD 90,000 - 150,000; Full time

Job summary

ROZOR in Hamilton, Ontario seeks a Robot Learning Engineer to develop learning-based systems for autonomy across ROZOR robots, focusing on RL/IL models for navigation in complex spaces and refining behaviors for cleaning or delivery tasks in dynamic indoor settings.

You will build training pipelines in simulation, design reward structures, and collaborate with cross-functional teams to integrate learned policies with perception and embedded systems, enabling robust and generalized robot behavior.

Qualifications

• Bachelor's or Master's in Robotics, AI, CS, or related field.
• Expertise in RL/IL or decision-making algorithms.
• Proficiency in PyTorch, TensorFlow, and modern simulation frameworks.
• Experience with ROS / ROS 2.
• Strong understanding of robot kinematics, navigation, and control systems.
• Strong foundation in optimization, probability, and ML.

Responsibilities

• Develop reinforcement / imitation learning algorithms for navigation, manipulation, and task execution.
• Build training pipelines in simulation for scalable robot learning.
• Create policies for obstacle avoidance and motion planning in dynamic environments.
• Design reward functions, curriculum strategies, and evaluation metrics.
• Analyze and reduce simulation-to-reality gaps.
• Collaborate with autonomy, perception, and embedded teams to integrate learned policies.
• Improve robustness and generalization of learned behaviors.

Skills

Reinforcement learning

Imitation learning

Decision-making algorithms

PyTorch

TensorFlow

Robotics simulation frameworks

ROS / ROS 2

Robot kinematics

Navigation

Control systems

Optimization

Probability

Machine learning

Education

Bachelor's or Master's in Robotics/AI/CS

Tools

PyTorch

TensorFlow

ROS/ROS 2

Robotics simulation frameworks

Job description

ROZOR in Hamilton, Ontario seeks a Robot Learning Engineer to develop learning-based systems for autonomy across ROZOR robots, focusing on RL/IL models for navigation in complex spaces and refining behaviors for cleaning or delivery tasks in dynamic indoor settings.

You will build training pipelines in simulation, design reward structures, and collaborate with cross-functional teams to integrate learned policies with perception and embedded systems, enabling robust and generalized robot behavior.