machine learning engineer
About the company
Mobio Interactive brings objective measurement to mental healthcare through its commercial platform, Dawn . As part of an Ontario Brain Institute (OBI) Centre for Analytics (CfA) project, we are developing and validating a multimodal AI framework using PyTorch and Latent Space Alignment. The goal is to map smartphone-derived physiological biomarkers to hospital-grade clinical telemetry (EEG/ECG) and psychiatric assessments within a shared 3D affective space covering Stress, Valence, and Arousal. We are seeking a Highly Qualified Personnel (HQP) candidate—a current graduate student or recent Master’s/PhD graduate—with a strong computational background and an interest in biomedical time-series data. Prior Computer Vision or NLP/LLM experience is not required.
Your tasks
- Data Preprocessing: Clean, filter, and harmonise clinical EEG, ECG/HRV, and psychometric datasets within the secure, PHIPA-compliant OBI Brain-CODE cloud workspace (Linux/Azure).
- Model Development and Training: Build, implement, and train PyTorch-based representation learning models to align clinical telemetry with smartphone-derived physiological features.
- Model Validation: Conduct clustering, cross-modal concordance, and statistical distance analyses to evaluate the alignment and consistency of the learned representations.
- Reproducible Research: Develop and document reproducible Jupyter Notebooks for a public OBI Methods Tutorial, presenting the project's methodology and analytical workflow.
Requirements
A core goal of this OBI-funded role is hands-on skill development. Candidates with strong Python and machine learning fundamentals who are motivated to learn neuroanalytics are encouraged to apply, even if they do not meet every criterion.
- Education: Current or recent Master’s/PhD student in Computer Science, Data Science, Biomedical Engineering, Computational Neuroscience, or a related field.
- Technical Skills: Proficiency in Python, including NumPy, Pandas, scikit-learn, and SciPy, with hands-on experience using PyTorch.
- Communication: Strong written English and familiarity with Jupyter Notebooks.
- Eligibility: Must be physically based in Ontario, Canada, and legally authorised to work in Canada.
- Nice to Have: Experience with biomedical time-series data (EEG/ECG/PPG), MNE-Python, NeuroKit2, or Linux/SSH.
What we offer
- Receive direct mentorship from Mobio’s Chief Scientist in clinical neuroinformatics and research methodology, and from the Computer Vision Scientist in technical modelling and pipeline integration.
- Gain hands-on experience working with clinical biomedical datasets in Ontario’s Brain-CODE platform.
- Contribute to reproducible research materials and receive authorship or credit on the public OBI Methods Tutorial, as appropriate.