AI Research Assistant: Multimodal Foundation Models

University Health Network; Toronto; On-site; CAD 31,000 - 39,000; Part time

Benefits offered by this job

HOOPP

Transit access

Flexible work environment

Development opportunities

Discounts

Job summary

University Health Network (UHN) invites applications for a Research Technical Assistant to support AI and multimodal foundation models at the PMCC. The role focuses on developing and evaluating state-of-the-art AI methods for biomedical data, working with genomics, EHRs, clinical text, and imaging to enable translational insights.

The successful candidate will collaborate with clinicians, AI scientists, and researchers, contributing to software, publications, and reproducible workflows in a

Qualifications

• Currently enrolled in an undergraduate, master's, or doctoral program in Computer Science, Biomedical Engineering, Computational Biology, Bioinformatics, Data Science, Artificial Intelligence, Statistics, Applied Mathematics, or related quantitative discipline.
• Strong programming in Python and experience with scientific computing or machine-learning workflows.
• Experience with deep learning frameworks such as PyTorch.
• Familiarity with foundation models, large language models, multimodal AI, or agentic AI frameworks such as LangChain or LangGraph is an asset.
• Solid understanding of machine-learning and deep-learning fundamentals

Responsibilities

• Assist in the development and evaluation of foundation models and multimodal AI methods for biomedical and healthcare applications.
• Contribute to the design and implementation of agentic AI systems, including single- and multi-agent workflows for reasoning, information integration, and decision support.
• Work with multimodal biomedical and clinical data, with a primary focus on genomics, electronic health records, clinical text, molecular data, and other structured and unstructured health data.
• Assist with the development and evaluation of methods for integrating longitudinal clinical and genomic information into unified AI representations.
• Develop, test, and optimize machine-learning and deep-learning methods using Python, PyTorch, and related frameworks.
• Participate in data preprocessing, representation learning, model training, benchmarking, and evaluation of AI models.
• Contribute to research on foundation-model adaptation, agent memory, reasoning, and multimodal learning for healthcare applications.
• Conduct literature reviews and remain current with developments in foundation models, large language models, multimodal AI, and autonomous/multi-agent AI systems.
• Assist with research experiments, analysis of results, preparation of figures, and interpretation of model performance.
• Contribute to research manuscripts, conference submissions, presentations, technical reports, and open-source research software.
• Follow software-engineering and reproducible-research best practices, including version control, documentation, experiment tracking, and reproducible computational workflows.
• Collaborate with AI researchers, staff scientists, computational scientists, clinicians, and other domain experts on interdisciplinary biomedical research projects.
• Participate actively in research meetings, project discussions, and regular progress reviews.
• Currently enrolled in an undergraduate, master's, or doctoral program in Computer Science, Biomedical Engineering, Computational Biology, Bioinformatics, Data Science, Artificial Intelligence, Statistics, Applied Mathematics, or a related quantitative discipline.

Skills

Python

PyTorch

LangChain

LangGraph

Linux

Education

Enrollment in Undergraduate/Master/PhD program in CS or related quantitative field

Tools

Git

SLURM

Linux

Job description

University Health Network (UHN) invites applications for a Research Technical Assistant to support AI and multimodal foundation models at the PMCC. The role focuses on developing and evaluating state-of-the-art AI methods for biomedical data, working with genomics, EHRs, clinical text, and imaging to enable translational insights.

The successful candidate will collaborate with clinicians, AI scientists, and researchers, contributing to software, publications, and reproducible workflows in a