Canva
Senior Research Engineer - Evaluations
Found: September 12, 2025
This role is based in San Francisco, CA with a hybrid work model.
Compensation:
$220,000 - $280,000/year
About the role:
Engineer AI agents to assess the quality and alignment of generative design models.
Responsibilities:
- Design and optimize an 'MLLM-as-a-Judge' evaluation system.
- Implement inference-time alignment techniques to improve model output quality.
- Establish a benchmarking process for foundation models.
- Analyze evaluation data and provide actionable insights.
- Collaborate with research scientists and ML engineers.
Requirements:
- Strong understanding of generative AI models and architectures.
- Experience with PyTorch and large-scale distributed model training.
- Familiarity with cloud environments, ideally AWS.
