Research Assistant I – STEM
The AI for Mental Health Program at St. Michael’s Hospital led by Dr. Venkat Bhat is currently looking for a Research Assistant (RA) . The AI for Mental Health Program offers novel psychopharmacological (e.g. IV ketamine/other anesthetic agents), neurostimulation (e.g. rTMS, emerging neurostimulation modalities) and digital therapeutic (mobile-based monitoring and interventions) interventions for Mood and other Disorders. The interdisciplinary program offers emerging and procedural interventions for psychiatric disorders, the research program aims to understand treatment-response with these emerging interventions. In addition to Dr. Bhat, the RA will have the opportunity to work with leading researchers in psychiatry, anesthesia, imaging, informatics and biomedical engineering. This position will focus on psychiatric clinical trials and digital mental health/artificial intelligence.
The AI for Mental Health Program is currently looking for a Research Assistant . The primary role of the Research Assistant I includes contributing to the comprehensive research and development lifecycle — from fine-tuning small language models and building API-integrated pipelines, to conducting rigorous performance evaluations and co-authoring scientific manuscripts. Candidates with hands-on experience in mental health, psychology, or neurology settings are especially encouraged to apply.
This is an entry-level role with potential for progression to level II and potentially Research Coordination.
Don’t meet every single requirement? Studies have shown that people in underrepresented communities are less likely to apply to jobs when they don’t meet every single qualification. We are dedicated to building an inclusive workplace, so if you’re excited about this role but your past experience doesn’t align perfectly, we still welcome you to apply.
Duties & Responsibilities:
Due to the variable nature of the position, this list is to be used as a guide only.
AI & Model Development
Fine-tune small language models (SLMs) such as Phi, Gemma, Mistral, and LLaMA variants for mental health-specific tasks including symptom assessment, psychoeducation, and conversational support.
Design and implement multi-step agentic workflows using frameworks such as LangChain and LangGraph.
Integrate and orchestrate AI capabilities via API calls to foundation model providers (e.g., Anthropic, OpenAI, Google, Hugging Face Inference API).
Develop prompt engineering strategies, retrieval-augmented generation (RAG) pipelines, guardrail, and tool-use patterns tailored to clinical and behavioural health contexts.
Performance Evaluation
Design evaluation frameworks and benchmarks to assess model accuracy, safety, clinical relevance, and alignment with mental health guidelines (e.g., DSM-5, PHQ-9, GAD-7).
Conduct red-teaming, bias audits, and hallucination assessments specific to mental health outputs.
Implement A/B testing and continuous evaluation pipelines to track model performance over time.
Document findings and translate results into actionable improvements for the engineering team.
Research & Manuscript Writing
Contribute to the design and execution of research studies, including literature reviews, methodology development, and IRB/ethics documentation.
Co-author peer-reviewed papers, conference submissions, and technical reports for PubMed indexed journals.
Synthesize findings from experiments and translate them into clear, rigorous scientific prose.
Maintain organized research logs, experiment tracking (e.g., MLflow, Weights & Biases), and reproducible code repositories.
Qualifications:
Undergraduate Degree or 1 year of relevant experience OR demonstrable equivalent combination of specialized education and experience.
TCPS CORE 2 is preferred (Completed within first 2 weeks of hire)
Bachelor's or Master's degree (or equivalent experience) in Computer Science, AI/ML, Biomedical Engineering, Cognitive Science, or a related field.
Proficiency in Python and experience with ML/NLP libraries including PyTorch, Hugging Face Transformers, and PEFT/LoRA for fine-tuning.
Demonstrated experience making API calls to LLM providers and building production-ready integrations.
Familiarity with evaluation methodologies for generative AI, including automated metrics and human-in-the-loop assessments.
Strong scientific writing ability with evidence of prior manuscript preparation, thesis writing, or published work.
Ability to work independently, manage multiple concurrent tasks, and communicate findings clearly to both technical and non-technical stakeholders.
Valued Assets:
Cloud & DevOps
Experience deploying ML workloads on cloud platforms such as AWS, Azure, or Google Cloud.
Familiarity with enterprise AI deployment practices including CI/CD pipelines, Docker/Kubernetes containerization, etc.
Experience with MLOps tools for experiment tracking, model versioning, and automated retraining pipelines.
Knowledge of data privacy and security requirements relevant to healthcare AI (PHIPA, PEDIPA, GDPR).
Mental Health, Psychology & Neurology
Academic or professional background in clinical psychology, psychiatry, counseling, neuroscience, or public mental health.
Familiarity with validated clinical instruments and psychometric scales (e.g., PHQ-9, GAD-7).
Understanding of therapeutic modalities relevant to AI-assisted interventions (e.g., CBT, DBT, motivational interviewing).
Experience with neuroimaging data, EEG/physiological signal processing, or digital phenotyping is a plus.
Prior work in a clinical, research, or peer support setting within mental health or neurology contexts is highly valued.
Unity Health Toronto is committed to creating an accessible and inclusive organization. We strive to provide a recruitment process that is barrier-free and in compliance with the Accessibility for Ontarians with Disabilities Act (AODA) and the Ontario Human Rights Code. We understand that you may require an accommodation at any stage of the recruitment process. When you are contacted, please inform the Talent Acquisition Specialist and we will work with you to meet your accommodation needs. We want to emphasize that all accommodation requests are handled with the utmost confidentiality, respecting your privacy and dignity.
