PhD Research Scientist Intern

Canva
San Francisco
Workplace: HybridInternshipUSD 157,000+ annuallyFunction: Data Science & Machine LearningEducation: phdSkills: ["Scientific rigor","Clear communication","Cross-team collaboration"]

Build and evaluate synthetic training data for Canva’s design agent, generating persona-conditioned conversations with tool calls and design-state changes. Combine persona-driven user-simulation with existing intent-labeling and clustering to keep synthetic outputs aligned with real usage. Measure realism gaps and distribution fit, collaborate with research and product teams on substitution for real data in evaluation, and share results with the research community.

Loading

Loading job details...

Preparing the role view and application actions.

FursaFursa
Canva
Canva
14 hours ago

PhD Research Scientist Intern

✓ Verified Job

Canonical indexed version, validated from employer's careers page.

Source: Company careers pageValidated by: Fursa AI
Last checked: 4 hours agoStatus: Live

Job Summary

Build and evaluate synthetic training data for Canva’s design agent, generating persona-conditioned conversations with tool calls and design-state changes. Combine persona-driven user-simulation with existing intent-labeling and clustering to keep synthetic outputs aligned with real usage. Measure realism gaps and distribution fit, collaborate with research and product teams on substitution for real data in evaluation, and share results with the research community.
Location: San Francisco
Workplace: Hybrid
Employment Type: Internship · 4 months
Job Function: Data Science & Machine Learning
Seniority: Intern level

Key Responsibilities

  • •Develop a synthetic data pipeline that produces persona-conditioned conversations for a design agent (user turns, tool calls, design-state changes).
  • •Apply persona-conditioned user-simulation techniques combined with an intent-labeling and clustering pipeline to keep synthetic outputs in distribution.
  • •Measure realism gap, distributional fit, and downstream evaluation quality using rigorous, reproducible validation (e.g., separability and agreement against ELO rankings).
  • •Partner with eval/labelling, research, and product teams to assess whether synthetic data can substitute for real user-generated content in evaluation.
  • •Contribute methodology and findings back to the research community via publication when results support it.

Pay and Benefits

Salary: USD 157,000 annually

Key Requirements

  • •Currently completing a PhD (ideally third year or later).
  • •Experience with human-AI interaction, conversational agents, or user-modeling research.
  • •Ability to design, run, and interpret machine-learning experiments with strong scientific rigor.
  • •Ability to develop research code for data processing, model training, and evaluation.
  • •Clear written and presentation communication, with strong cross-team collaboration.
Experience:Generative AIAI researchConversational agentsHuman-AI interaction
Education:PhD / Doctorate
Skills:Scientific rigorClear communicationCross-team collaboration
Languages:English
Tech Stack:Machine learningGenerative AISynthetic dataConversational agentsUser modelingData processingModel trainingEvaluation

Company Brief

Canva
Canva is an online visual communications platform that provides drag-and-drop design tools, templates, and media assets for creating presentations, social media graphics, videos, documents and more, serving individuals and large enterprises globally.
Industry: Enterprise Software
Company Size: Enterprise (1,001+ employees)
Revenue: USD 1B+
Growth: Scaleup
Valuation: Decacorn (USD 10B+)
Funding: Series E+
Headquarters: Surry Hills, New South Wales, Australia
Founded: 2012
Glassdoor
Glassdoor: 3.9
WebsiteLinkedInGlassdoor