Research Scientist – Video Generation

This position is open to candidates at all experience levels, including experienced candidates, 2025 and 2026 graduates, as well as internship opportunities. The role is based in Beijing. We welcome your application and look forward to having you on board!

About the Role

• This Research Scientist role focuses on advanced video generation—pushing the boundaries of AI-driven storytelling. You'll join a close-knit team of researchers and engineers focused on developing new capabilities in audio-visual synthesis, dubbing, and controllable video creation. With access to world-class GPU infrastructure and real-world data, your work will directly power features used by millions across the globe.

What you'll do (responsibilities)

• You will work on multi-person video dubbing and audio-driven video generation
• You will build advanced diffusion or transformer-based models for controllable video synthesis
• You will conduct research on visual storytelling and multi-layer video composition.
• You will collaborate with Canva’s global AI teams to bring research into products used by creators worldwide

Qualifications

What We’re Looking For

• Strong background in computer vision, multimodal learning, or generative modeling
• Hands-on experience with video generation, audio-visual alignment, or speech-driven animation
• Proficiency in PyTorch and large-scale GPU training
• Publications at top venues (CVPR, ICCV, NeurIPS, ICLR, etc.) or equivalent applied research experience

Additional Information

Why Canva

• Access to large-scale H200 GPU clusters
• Work with unique real-world design and video datasets
• Collaborate with top global researchers advancing creative intelligence