Staff Machine Learning Engineer, Sensors – Intelligence

This is a fully remote position, open to applicants in Canada, +1 more country.

📋 Description

• Spearhead the research and development of multimodal foundation models utilizing synchronized sensor, video, and physiological signals.

• Create self-supervised pretraining methodologies for extensive, diverse time-series data.

• Formulate sensor-fusion techniques to tackle challenges related to sampling rates, sensor noise, synchronization errors, and absent modalities.

• Enhance 2D and 3D human pose estimation, motion reconstruction, and movement quality evaluation.

• Construct models that estimate joint kinematics, load, effort, fatigue, and recovery.

• Establish evaluation frameworks and benchmarks to ensure generalization across various users, body types, movements, and equipment conditions.

• Validate model outputs against motion capture data, force plates, and clinical-grade physiological sensors.

• Collaborate with ML Engineering to modify research models for deployment on cloud platforms and for on-device inference.

• Publish findings in leading venues and represent Tonal through academic papers, presentations, and partnerships.

• Mentor scientists and engineers while contributing to the scientific direction of the Sensors & Intelligence team.

⛳️ Requirements

• Demonstrated independent researcher with a proven history of publications in relevant academic venues.

• PhD in Computer Science, Electrical Engineering, Biomedical Engineering, Biomechanics, or a similar field strongly preferred; equivalent research experience and publication achievements are also welcomed.

• In-depth knowledge in sensor fusion, biomechanics, physiological signal modeling, or computer vision concerning human motion.

• Practical experience in designing, training, and troubleshooting transformer architectures, including attention mechanisms, positional encoding for time series, and multimodal tokenization.

• Proven capability to train and develop machine learning models end-to-end using PyTorch, JAX, or TensorFlow.

• Solid foundation in signal processing, encompassing filtering, calibration, time alignment, and frequency-domain analysis.

• Strong experimental design skills, including conducting ablation studies and ensuring no leakage across subjects.

• Ability to independently identify and advance ambiguous research challenges towards measurable product outcomes.

• Effective communication of complex research findings to scientists, engineers, and product leaders.

• Legally authorized to work in the US and/or Canada.

🏝️ Benefits

• Equity

• Health insurance

• Retirement savings benefits

• Life insurance

• Disability benefits

• Flexible paid time off

• Parental leave

• Other additional benefits