Senior RL Engineer: Simulation & Policy Design

NBCUniversal; Montreal (administrative region); On-site; CAD 90,000 - 140,000; Full time

Job summary

NBCUniversal is seeking a Reinforcement Learning Engineer to design robust virtual training environments for autonomous agents. You will build and maintain 2D/3D simulators (Unity/Unreal/Isaac Sim) and craft reward structures and policies aligned with product goals and safety constraints.

Ideal candidates have graduate degrees in AI/CS/robotics, strong Python/Git skills, and experience with RL frameworks and physics engines.

Qualifications

• Education: Graduate degree in Robotics, Computer Science, AI, or related field.
• Professional Experience: Proven experience as an RL Engineer or Research Engineer in a fast-paced environment.
• Industry Context: Prior experience in robotics, smart grids, precision agriculture, game development, or aerospace.

Responsibilities

• Cross-Functional Coordination with ML and Annotation engineers and TPMs to specify data, simulation, and training needs.
• Environment Design: Build high-fidelity 2D/3D simulation environments (Unity, Unreal, Isaac Sim) for RL training.
• Reward Engineering: Design and tune reward functions aligning agent behavior with product goals and safety constraints.
• Algorithm Implementation: Develop and optimize RL algorithms (PPO, SAC, Offline RL) for high-dimensional observations.
• Sim-to-Real Strategy: Address reality gap with domain randomization and adaptation techniques.

Skills

Python

Git

Unix shell

Reinforcement Learning

MDP

Education

Graduate degree (Master’s or PhD) in Robotics, CS, AI, or related field

Tools

Ray RLlib

Stable Baselines3

CleanRL

Unity

Unreal

Isaac Sim

MuJoCo

Bullet

Jira

Confluence

Slack

Job description

NBCUniversal is seeking a Reinforcement Learning Engineer to design robust virtual training environments for autonomous agents. You will build and maintain 2D/3D simulators (Unity/Unreal/Isaac Sim) and craft reward structures and policies aligned with product goals and safety constraints.

Ideal candidates have graduate degrees in AI/CS/robotics, strong Python/Git skills, and experience with RL frameworks and physics engines.