Machine Learning Engineer - Motion Capture
Peripheral Labs
Who we are:
Peripheral is developing spatial intelligence, starting in live sports and entertainment. Our models generate interactive, photorealistic 3D reconstructions of sporting events, building the future of live media. We’re solving key research challenges in 3D computer vision, creating the foundations for the next generation of robotic perception and embodied intelligence.
We’re backed by Tier-1 investors and working with some of the biggest names in sports. Our team includes top robotics and machine learning researchers from the University of Toronto, advised by Dr. Steven Waslander and Dr. Igor Gilitshenski.
Our team is ambitious and looking to win. We’re seeking a Machine Learning engineer to develop our motion capture models through synthetic data curation, model training, and inference-time optimization.
What you’ll be doing:
Developing our data flywheel to autolabel and generate synthetic data,
Improving our motion capture accuracy by fine-tuning existing models on our domain,
Optimizing inference time through model distillation and quantization,
What we’d want to see:
Prior experience with 3D computer vision and training new ML models,
Strong understanding of GPU optimization methods (Profiling, Quantization, Model Distillation),
Proficiency in Python and real-time ML inference backends,
Ways to stand out from the crowd:
Previous experience in architecting and optimizing 3D computer vision systems,
Strong understanding of CUDA and Kernel programming,
Familiarity with state-of-the-art research in VLMs,
Top publications at conferences like NeurIPS, ICLR, ICML, CVPR, WACV, CoRL, ICRA,
Why join us:
Competitive equity as an early team member.
$80-120K CAD + bonuses, flexible based on experience.
Exclusive access to the world’s biggest sporting events and venues,
Work on impactful projects, developing the future of 3D media and spatial intelligence.