Staff Research Engineer - Audio & Video AI

Location: Team Engineering

Team: Country Sydney / Australia

About the Role

As a Staff Research Engineer - Audio & Video AI at Canva, you'll shape the future of intelligent design by pioneering multimodal AI that brings audio and video magic to millions of creators. Sitting at the intersection of cutting-edge research, large-scale ML engineering, and technical leadership, you'll transform novel ideas into production-ready features that empower everyone to design. Based in our vibrant Sydney flagship campus, you'll collaborate with world-class researchers, designers, and engineers in a culture that celebrates curiosity, bold innovation, and delightful user experiences. Your days will blend hands-on deep learning with strategic impact: partnering with scientists to productionize generative breakthroughs in audio/video, architecting scalable ML systems, and optimizing for Canva's high-velocity ecosystem. From rapid prototyping novel diffusion models to building TB-scale datasets and distributed training pipelines, you'll own the journey from research spark to magical design tool. As a technical leader, you'll mentor teams, influence roadmaps, and set excellence standards that elevate Canva's AI across the organization. You'll thrive if you bring rigorous engineering to creative challenges—mastery in PyTorch, experience scaling massive models, and a passion for multimodal AI that powers real-world creativity. Don't worry if you don't check every box; at Canva, we value diverse backgrounds, relentless curiosity, and the drive to learn. Join our collaborative, design-obsessed team to redefine how the world designs, with equity, flexible benefits, and endless opportunities for growth in Sydney's sunny innovation hub.

Key Responsibilities

• Partner with Research Scientists to productionize advances in audio, video, and multimodal generation
• Lead rapid prototyping of novel architectures, algorithms, and training approaches
• Design and own end-to-end ML systems for generative and analytic tasks in audio/video domains
• Build and scale production-grade ML pipelines including training, fine-tuning, deployment, and monitoring
• Optimize large-scale models for efficiency, latency, and throughput in distributed environments
• Establish high-quality datasets and annotation pipelines for multimodal learning at scale
• Collaborate cross-functionally with product, engineering, design, and research teams
• Provide technical leadership, mentor engineers/researchers, and influence ML strategy
• Drive integration of AI capabilities into Canva's design platform for millions of users
• Set engineering best practices and raise technical standards across the ML organization
• Align ML roadmaps with Canva's mission to empower everyone to design

Required Qualifications

• Deep expertise in state-of-the-art generative models (e.g., diffusion models, GANs, transformers) with proven application to real-world products
• Experience leading large-scale distributed training across hundreds of GPUs, including infrastructure and performance tradeoffs
• Strong foundation in ML and deep learning with mastery in PyTorch for model implementation, optimization, and scaling
• Hands-on experience with large-scale audio or video datasets (TB/PB scale), including preprocessing and representation learning
• Demonstrated engineering rigor in building production ML systems with clean code, modular architectures, and observability
• Track record of end-to-end ML system ownership from research prototyping to deployment and monitoring
• 5+ years of experience in ML engineering or research engineering roles at scale
• Bachelor's or higher degree in Computer Science, Electrical Engineering, or related field

Preferred Qualifications

• Experience in multimodal AI combining audio, video, and text for generative applications
• Prior work at a design or creative tools company building user-facing AI features
• Contributions to open-source ML projects or publications in top AI conferences
• Familiarity with Canva's design ecosystem or similar creative platforms
• Leadership in mentoring ML engineers and researchers
• Experience optimizing models for low-latency inference in consumer applications

Required Skills

• PyTorch mastery for complex model development
• Distributed training at massive scale (100s GPUs)
• Generative AI models (diffusion, GANs, transformers)
• Audio/video data processing and representation learning
• Production ML pipelines (CI/CD, monitoring, evaluation)
• Model optimization for inference efficiency
• Dataset curation and annotation at scale
• Cross-functional collaboration and stakeholder alignment
• Technical leadership and mentorship
• Pragmatic problem-solving in ambiguous environments
• Clean, maintainable code and modular architectures
• Strong communication and knowledge-sharing
• Perceptual evaluation methodologies for media AI
• End-to-end system ownership mindset
• Design-thinking for user-centric AI features
• Rapid prototyping and iteration

Benefits

• Equity packages to share in Canva's success
• Inclusive parental leave policy supporting all parents and carers
• Annual Vibe & Thrive allowance for wellbeing, social connection, and office setup
• Flexible leave options to recharge and support personal needs
• Flagship Sydney campus with collaborative, design-focused environment
• Delightful moments of magic, connectivity, and fun woven into work life
• Opportunities to work on world-class AI shaping creative experiences
• Comprehensive health and wellness programs
• Global team with diverse, inclusive culture

Canva is an equal opportunity employer.