Staff Machine Learning Engineer, Sensors & Intelligence

CanadaRemoteFull-timelead

Machine Learning EngineerData

1 views0 saves0 applied

Apply Now

Quick Summary

Requirements Summary

Proven ability to conduct independent research, supported by a strong publication record at relevant conferences or journals, such as NeurIPS, ICML, ICLR, CVPR, ICCV, ECCV, IMWUT/UbiComp, CHI,

Technical Tools

Machine Learning EngineerData

This is an opportunity to shape the next generation of machine learning models for understanding human movement, performance, and physiology. You will work at the intersection of machine learning, biomechanics, physiological signal processing, and computer vision, leveraging large-scale multimodal workout data. Your research will help create general-purpose representations of how people move, exert effort, experience fatigue, and adapt over time. These capabilities will support intelligent coaching, movement feedback, personalized training progression, and advanced musculoskeletal health applications. Combining foundational research with real-world product development, you will influence scientific direction, publish impactful work, and collaborate with engineering teams to bring new capabilities to users. You will also mentor colleagues and help shape the growth of a multidisciplinary research team.

Lead research and development of multimodal foundation models that learn from synchronized sensor, video, and physiological data.
Design self-supervised pretraining approaches for long, heterogeneous time-series datasets.
Develop sensor fusion techniques that account for different sampling rates, sensor noise, synchronization errors, and missing data modalities.
Advance computer vision methods for 2D and 3D human pose estimation, motion reconstruction, and movement quality assessment.
Build machine learning models to estimate biomechanical and physiological measures, including joint kinematics, physical load, effort, fatigue, and recovery.
Establish evaluation frameworks and benchmarks to assess model generalization across different users, body types, movements, and equipment conditions.
Validate model predictions against high-quality reference measurements, such as motion capture systems, force plates, and clinical-grade physiological sensors.
Partner with ML Engineering teams to adapt research models for cloud deployment and on-device inference.
Publish research regularly at leading academic and industry venues, and represent the team through presentations, collaborations, and participation in the research community.
Mentor scientists and engineers while helping define the scientific priorities and technical direction of the Sensors & Intelligence team.

Requirements

~1 min read

Proven ability to conduct independent research, supported by a strong publication record at relevant conferences or journals, such as NeurIPS, ICML, ICLR, CVPR, ICCV, ECCV, IMWUT/UbiComp, CHI, IEEE TBME, or the Journal of Biomechanics.
PhD in Computer Science, Electrical Engineering, Biomedical Engineering, Biomechanics, or a related discipline is strongly preferred. Candidates with equivalent research experience and a strong publication record are also encouraged to apply.
Deep expertise in at least one of the following areas: sensor fusion, biomechanics, physiological signal modeling, or computer vision applied to human motion.
Practical experience designing, training, and debugging transformer architectures, including attention mechanisms, time-series positional encoding, and multimodal tokenization.
Strong end-to-end machine learning development skills using PyTorch, JAX, or TensorFlow, covering data curation, model training, experimentation, and analysis.
Solid understanding of signal processing techniques, including filtering, calibration, temporal alignment, and frequency-domain analysis.
Strong experimental design capabilities, including building ablation studies, preventing subject-level data leakage, and distinguishing meaningful performance improvements from statistical noise.
Ability to independently identify, structure, and lead ambiguous research challenges toward measurable product outcomes.
Excellent communication skills, with the ability to explain complex technical findings to research, engineering, and product stakeholders.

Experience pretraining or scaling foundation models, including distributed training and an understanding of model scaling behavior.
Familiarity with musculoskeletal modeling tools such as OpenSim, MuJoCo, or MyoSim, or experience with physics-informed machine learning.
Experience working with wearable physiological signals, including IMU, PPG, heart rate variability (HRV), or EMG data.
Knowledge of personalization methods, few-shot adaptation, or uncertainty quantification.
Experience with model compression or efficient architectures for edge and on-device deployment.
Background in human-subject research, data collection protocols, or scientific validation studies.
Interest in fitness, sports science, or digital health.

Candidates are encouraged to apply even if they do not meet every listed qualification. Diverse backgrounds, research perspectives, and transferable experience are valued.