PhD Research Scientist Intern - Edge AI

Canva
Vienna
Workplace: HybridInternshipFunction: Data Science & Machine LearningEducation: phdSkills: ["Documentation","Planning","Measurement","Iteration","Collaboration"]

Work on Canva’s Video Storytelling models to optimize and deploy video-capable vision-language models for efficient on-device inference. You’ll survey VLMs, apply techniques like quantization, pruning, distillation, and fine-tuning, and deploy through an on-device inference library. The internship includes end-to-end research, benchmarking against reference pipelines, evaluating caption quality versus server-side bars, and publishing findings via a paper and patent filing.

Loading

Loading job details...

Preparing the role view and application actions.

FursaFursa
Canva
Canva
5 hours ago

PhD Research Scientist Intern - Edge AI

✓ Verified Job

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

Source: Company careers pageValidated by: Fursa AI
Last checked: 5 hours agoStatus: Live

Job Summary

Work on Canva’s Video Storytelling models to optimize and deploy video-capable vision-language models for efficient on-device inference. You’ll survey VLMs, apply techniques like quantization, pruning, distillation, and fine-tuning, and deploy through an on-device inference library. The internship includes end-to-end research, benchmarking against reference pipelines, evaluating caption quality versus server-side bars, and publishing findings via a paper and patent filing.
Location: Vienna
Workplace: Hybrid
Employment Type: Internship · Permanent
Job Function: Data Science & Machine Learning
Seniority: Intern level

Key Responsibilities

  • •Survey candidate video-capable VLMs and select the best starting point.
  • •Apply model optimization and architecture improvements for on-device deployment (e.g., quantization, pruning, distillation, fine-tuning for caption placement).
  • •Deploy the model to real high-traffic mobile hardware via Canva’s on-device inference library and iterate based on on-device measurements.
  • •Run comparative evaluation against alternative optimization paths and perform human evaluation against the server-side caption quality bar.
  • •Document and publish results, including a patent filing and a paper publication with findings and viability mapping for current consumer hardware.

Key Requirements

  • •Current enrolment in a PhD in ML, CS, or a related field, with first-author papers in venues such as CVPR, NeurIPS, ICCV/ECCV, ICLR, or ICML.
  • •Strong Python and hands-on PyTorch experience, including training and fine-tuning vision-language models.
  • •Ability to understand modern vision-language and multimodal architectures and reproduce results from recent papers.
  • •Experience with optimisation methods such as quantisation, pruning, or distillation, including trade-offs in accuracy.
  • •Experience deploying models on-device/at the edge using runtimes such as Core ML, LiteRT/TFLite, ONNX Runtime, or ExecuTorch, within memory and latency budgets.
Experience:Edge AIEfficient MLMultimodal models
Education:PhD / Doctorate in ML, CS, or a related field
Skills:DocumentationPlanningMeasurementIterationCollaboration
Languages:English
Tech Stack:PythonPyTorchQuantizationPruningDistillationCore MLLiteRTTFLiteONNX RuntimeExecuTorchGemmaQwen-VLSmolVLMMiniCPM-VVision-language modelsMultimodal architectures

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