Sports Computer Vision Engineer
$165,000–$200,000 year
RemoteCanada
Job Summary
Build and train computer vision models for sports video, covering player and ball detection, multi-object tracking, pose estimation, and event recognition. Own the experimentation loop from hypothesis to ablation, designing evaluation metrics and failure taxonomies to drive measurable improvements. Prototype modern architectures including transformers and temporal models while optimizing data efficiency through augmentation and sampling strategies. Collaborate on dataset design, labeling schemas, and production pipelines to ensure reliable batch and stream inference. Add quality gates for reproducibility and automated regression detection to support scalable deployment.
Required Qualifications
- Strong applied CV experience with hands-on model development (not just running existing repos)
- Solid PyTorch skills: training loops, debugging, data pipelines for vision workloads, DDP basics
- Comfort with video CV fundamentals: occlusion, identity switches, temporal consistency, calibration, domain shift
- Strong Python engineering and a bias toward measurable outcomes
Desired Qualifications
- Sports video CV or adjacent domains (multi-agent tracking, pose, crowded scenes)
- Experience with video tooling (FFmpeg), efficient dataset formats (WebDataset/shards), or streaming/batching to GPUs
- MLOps/production experience: model packaging, CI for training/eval, serving (Triton/TorchServe), monitoring
Hiring someone like this?
Get your role in front of qualified candidates on Sorce.