Research Engineer - Evaluations - Remote

Overview

Join the team redefining how the world experiences design. At Canva, our mission is to empower the world to design. To ensure our generative AI models are truly helpful, we are seeking a talented Research/Machine Learning Engineer to build our next-generation evaluation system by leveraging automatic evaluations.

In Short

• Engineer sophisticated AI agents for evaluating generative design models.
• Focus on building practical systems for effective research application.
• Work on agentic evaluation systems using Multimodal Large Language Models.
• Implement inference-time alignment techniques to improve model outputs.
• Establish a benchmarking process for quality understanding models.
• Analyze evaluation data for actionable insights.
• Collaborate with research scientists and ML engineers.
• Translate research into production-ready engineering solutions.
• Experience with large-scale model training across GPUs.
• Work in cloud environments, ideally AWS.

Requirements

• Strong understanding of generative AI models and architectures.
• Ability to create data-driven evaluation methodologies.
• Experience managing large-scale distributed model training.
• Proficient in machine learning and optimizing PyTorch codes.
• Disciplined coding practices with experience in code reviews.
• Familiarity with evaluation libraries and frameworks is a plus.

Benefits

• Equity packages to share in the company's success.
• Inclusive parental leave policy for all parents and carers.
• An annual Vibe & Thrive allowance for various personal needs.
• Flexible leave options to recharge and support personal well-being.