Reinforcement Learning Engineer (Full-Time) - Humanoid Robot

AXIBO INC; Cambridge; On-site; CAD 60,000 - 80,000; Full time

Job summary

A robotics company is seeking a Reinforcement Learning Engineer to develop and deploy machine learning systems for humanoid and legged robots. This role involves working with control theory and deep learning to create adaptive behaviors, requiring a strong background in machine learning and robotics. The position is full-time, located in Cambridge, Ontario, with competitive salary and health insurance provided.

Benefits

Competitive salary

Health insurance

Regular performance evaluations

Qualifications

• 2+ years of experience applying deep reinforcement learning.
• Proficiency in Python and C++ programming.
• Experience in Linux development environments.

Responsibilities

• Develop reinforcement learning agents for robotic control tasks.
• Implement learning architectures using policy gradient methods.
• Work with hardware teams to deploy trained policies.

Skills

Deep reinforcement learning

Machine learning fundamentals

Control theory

Debugging skills

Independent experiment management

Education

Bachelor's or Master’s degree in Computer Science, Engineering, Robotics

Tools

PyTorch

JAX

TensorFlow

Linux

GPU-based training

Job description

About AXIBO

AXIBO is a robotics company pioneering the design, prototyping, and manufacturing of advanced robotic systems—all under one roof. We build everything in-house and take pride in delivering robust, reliable products that power automation across industries. Our fast-paced environment demands high levels of precision, organization, and execution—not just in engineering, but across all functions.

Position Overview

As a Reinforcement Learning Engineer, you will develop and deploy machine learning systems that enable intelligent behaviors in our humanoid and legged robots. You\'ll work at the intersection of control theory, deep learning, and robotics—helping close the loop between simulation and reality to bring adaptive behaviors into real-world machines.

Key Responsibilities

• Develop reinforcement learning agents for robotic control tasks such as locomotion, manipulation, and dynamic balance
• Implement learning architectures using policy gradient methods, actor-critic frameworks, and off-policy algorithms (e.g., PPO, SAC, TD3)
• Build reward functions, curriculum learning strategies, and simulation environments tailored for real-world transfer
• Design multi-agent training pipelines, including distributed rollouts, experience replay, and adaptive difficulty scaling
• Interface with Isaac Gym, Mujoco, Brax, and custom physics simulators to run large-scale experiments
• Work with hardware and firmware teams to deploy trained policies to embedded or real-time environments
• Design diagnostic tools and visualization dashboards to monitor training progress and system behavior
• Apply domain randomization, sim2real techniques, and sensor noise modeling to enhance policy robustness
• Maintain code quality through version control, testing, and modular design
• Stay current with academic literature and integrate novel RL methods as appropriate

Required Skills and Qualifications

• Bachelor\'s or Master’s degree in Computer Science, Engineering, Robotics, or a related field
• 2+ years of hands-on experience applying deep reinforcement learning to simulation or robotic control tasks
• Strong grasp of machine learning fundamentals and control theory
• Proficiency with PyTorch, JAX, or TensorFlow
• Programming experience in Python and C++
• Deep understanding of policy optimization, generalization, and environment design
• Experience working in Linux development environments and with GPU-based training pipelines
• Excellent debugging skills across ML, software, and hardware stacks
• Ability to independently manage experiments and rapidly iterate on model architectures

Preferred Experience (Bonus)

• Deployment of RL systems to real-world robots, especially legged or humanoid platforms
• Contributions to open-source RL frameworks or robotics middleware (e.g., ROS, Isaac ROS)
• Experience with imitation learning, behavior cloning, or inverse reinforcement learning
• Prior research/publications in reinforcement learning, multi-agent systems, or robotic control
• Familiarity with low-level robot interfaces, sensor fusion, or control loop tuning
• Knowledge of real-time systems, embedded software, or custom actuator control

Job Details

• Location: Cambridge, Ontario
• Work Environment: In-person (on-site at our Waterloo facility)
• Type: Full-time
• Compensation: Competitive salary (based on experience)
• Health Insurance: Provided
• Growth: Regular performance evaluations with potential for salary increases and stock option participation