Senior Research Scientist - Reinforcement Learning, MoEs

This role is based in Vienna, Austria.

About the role:

You’ll drive research directions and play a leading role in hands‑on work across the agent stack—from reward design and policy optimization to planning, memory, and tool orchestration.

Responsibilities:

• Develop agent systems for real tasks in design, vision, and language.
• Scale post-training and RL across distributed systems (PyTorch).
• Contribute to the research agenda for RL/agentic systems aligned with Canva’s product goals.
• Build reward models and learning loops.
• Develop simulation and sandbox tasks that surface failure modes.
• Help align on rigorous evaluation for agents.
• Collaborate with product, design, safety, and platform teams.
• Mentor teammates and present findings internally.

Requirements:

• Depth in implementing and post-training MoEs/LLMs/VLMs/Diffusion models.
• Strong experience with experimental design and reproducibility.
• Fluency in Python and PyTorch.
• Hands-on experience with policy optimization and reward modeling.
• Experience with large-scale training and cloud multimodal tooling.