Lead multimodal machine learning research at Tonal, using sensor fusion and computer vision to model human movement and physiology remotely from Toronto. The role is remote-first.
As a Staff Machine Learning Engineer, you will build foundation models from synchronized sensor, video, and physiological signals. You will shape research directions, run rigorous experiments, and translate findings into adaptive coaching, form feedback, personalized progression, and musculoskeletal health models. You will also publish and mentor across the Sensors & Intelligence team.
Key Responsibilities: Lead multimodal foundation model research and development
Design self-supervised pretraining for heterogeneous time series
Develop sensor fusion for noise, alignment, and missing modalities
Advance 2D and 3D human pose estimation methods
Adapt research models for cloud and on-device inference
Requirements: Robust publication record as an independent researcher
Deep expertise in sensor fusion, biomechanics, physiology, or computer vision
Hands-on transformer architecture design, training, and debugging
End-to-end PyTorch, JAX, or TensorFlow experience
Strong signal processing and rigorous experimental design grounding
Shape Tonal’s movement intelligence through foundation models, sensor fusion, computer vision, and rigorous machine learning research.