Agentic Reinforcement Learning Engineer

Advance agentic reinforcement learning research from Canada with PyTorch and multimodal AI expertise in a fully remote setting. Build intelligent systems for complex challenges.

This AI Research Engineer role focuses on next-generation intelligent systems and adaptive behavior. You will develop reinforcement learning algorithms for multimodal models, scalable training infrastructure, reward strategies, and evaluation frameworks. Research spans images, audio, text, diffusion-based generative models, autoregressive architectures, and unified multimodal systems.

Key Responsibilities:

• Develop reinforcement learning algorithms for multimodal AI models
• Design distributed infrastructure for scalable reinforcement learning training
• Optimize rewards, stability, alignment, and reward-hacking prevention
• Create simulation environments and multimodal training datasets
• Conduct experiments, benchmarks, evaluations, and research publications

Requirements:

• Master's degree in Computer Science or related discipline
• Strong AI research record with leading conference publications
• Large-scale multimodal reinforcement learning experimentation experience
• Advanced PyTorch and modern deep learning framework proficiency
• Experience building end-to-end reinforcement learning pipelines

Apply agentic reinforcement learning, multimodal modeling, PyTorch, and rigorous experimentation to improve autonomous AI systems remotely.