Senior Machine Learning Engineer, Multimodal Data

• Own the data foundations that power Canva's multimodal agent research — pipelines, datasets, and tooling that turn ambitious research ideas into trainable reality. • Join a cutting-edge post-training team developing new multimodal agentic systems and scalable training and evaluation loops. • Own the full data lifecycle: collection, curation, preprocessing, quality assurance, and delivery into training pipelines. • Work with significant autonomy over how data problems get solved, while aligning with the team on what matters most. • Based at Canva's London campus with hybrid working.

• Strong software engineering skills in Python, with experience building production-grade data pipelines and ML DevOps. • Practical experience with prompt engineering — designing, testing, and refining prompts for reliable LLM/VLM outputs. • Experience with ML data workflows: large-scale data processing and loading (e.g. Ray), data versioning, and training format considerations like tokenisation, batching, and sharding. • Hands-on experience with data pipelines for large-scale distributed ML training runs. • Familiarity with annotation tooling and human-in-the-loop data collection, such as Label Studio or internal systems. • Understanding of what "good data" looks like for LLM/VLM fine-tuning and the ability to anticipate downstream issues. • Experience loading and writing large datasets to and from cloud infrastructure (AWS) and distributed storage systems. • Strong communication skills, able to work with researchers to scope ambiguous problems into actionable plans.

• Experience with preference data collection for RLHF or reward modelling. • Familiarity with multimodal data such as image-text pairs, video, or design assets. • Experience building synthetic data generation pipelines using LLMs. • Background in data quality metrics and monitoring systems. • Contributions to dataset releases or benchmarks in the ML community.

• Design and build data pipelines for agent training: collection, filtering, deduplication, formatting, and versioning across text, image, and multimodal sources. • Build and maintain infrastructure for efficient data loading, storage, and retrieval at scale. • Collaborate with research scientists to translate research requirements into concrete data specifications, iterating as experiments reveal new needs. • Create evaluation datasets and benchmarks that surface real failure modes. • Develop tooling for dataset construction, including human annotation workflows, synthetic data generation, and preference data collection for RLHF/DPO-style training. • Own data quality — build validation frameworks, monitor for drift and contamination, and establish standards that make datasets trustworthy and reproducible. • Document datasets thoroughly: provenance, known limitations, intended use cases, and versioning history. • Implement comprehensive test coverage for data pipelines and ML workflows. • Elevate codebase quality through code reviews, refactoring, and establishing sustainable engineering best practices.

• Equity packages so Canva's success is shared with you. • Inclusive parental leave policy that supports all parents and carers. • Annual Vibe & Thrive allowance for wellbeing, social connection, and office setup. • Flexible leave options to recharge and support you personally. • Virtual interview process.

Canva is on a mission to empower the world to design, building AI that feels magical and lands real impact for millions of people. Its cutting-edge post-training team develops new multimodal agentic systems, building scalable training and evaluation loops that turn research breakthroughs into delightful product features.