PhD Research Scientist Intern - Reinforcement Learning, Images

📋 Description

• Internship with Canva's AI team on a live, industry-scale project
• 16 weeks, starts in September
• Hands-on with real data, production infra, and deadlines
• Collaborate with researchers, engineers, and product teams toward production
• Begin on novel parts from week one due to established groundwork

🎯 Requirements

• PhD student, ideally third year or later
• Strong diffusion or flow-matching background
• Hands-on policy-gradient RL for generative models (GRPO, PPO, DPO or similar)
• Experience fine-tuning VLMs (e.g., LoRA) and prompts/rubrics for evaluation
• Reward modelling, preference optimisation, pseudo-labelling, distillation
• Ability to read and reproduce recent papers; clear communication in writing/presentations

🎁 Benefits

• London-based hybrid role with option to work from home
• Work on impactful AI projects used by millions
• Mentorship from four mentors and cross-team collaboration
• Opportunity to publish results and contribute to Canva's research roadmap