Drive innovation as a Reinforcement Learning Engineer with Preference Model. This role focuses on creating high-quality RL environments targeting real-world application complexities.

As part of the Low Level/Kernels Capabilities team, you’ll merge your engineering and research skills to build RL tasks that push the boundaries of ML. Your work will involve designing and scoring low-level kernels while ensuring robustness against unintentional model exploits. This unique setting presents challenges in hardware optimization and task design.

Key Responsibilities:



• Create kernel-focused RL environments for various tasks • Develop reliable scoring metrics that prevent manipulation • Identify key RL environments for development • Design robust performance measurement systems • Own and implement projects from conception to deployment

Requirements: • Expertise in low-level systems engineering (C/C++/CUDA) • Proven record in Python for production-level coding • Hardware optimization understanding and application • Experience with kernel development and profiling • Ability to independently manage complete project cycles

Bring your skills in kernel design to help revolutionize machine learning at Preference Model. #J-18808-Ljbffr

📌 Reinforcement Learning Engineer at Preference Model (Winnipeg)
🏢 Preference Model
📍 Winnipeg