Sports Computer Vision Engineer

This is a fully remote position, open to applicants in United States.

📋 Description

• Develop and train computer vision models specifically for sports video, encompassing player/ball detection, multi-object tracking, pose/keypoints recognition, event/action identification, and identity association (re-ID).

• Take ownership of the experimentation process from hypothesis formulation through ablation studies and error analysis to achieve measurable enhancements.

• Design and sustain evaluation metrics tailored to tasks, dataset slices, and failure taxonomies.

• Enhance data efficiency via augmentations, sampling techniques, label noise management, and weak/self-supervision when beneficial.

• Prototype and refine transformer-based detection/tracking systems, temporal models, and multi-task architectures.

• Collaborate on the design of datasets and labeling, including formats, schemas, tools, and version control.

• Assist in productionizing models by implementing packaging, batch/stream inference methods, throughput/latency trade-offs, and robustness evaluations.

• Establish quality gates to ensure reproducibility, automated evaluations, and regression monitoring.

• Work closely with data and platform team members to guarantee reliable model deployment.

⛳️ Requirements

• Extensive practical experience in applied computer vision with direct model development (beyond merely executing existing repositories).

• Proficient in PyTorch: adept in training loops, debugging, and data pipelines for vision tasks, along with basic knowledge of DDP.

• Familiarity with fundamental video computer vision concepts, including occlusion, identity changes, temporal consistency, calibration, and domain shifts.

• Strong Python programming skills with a focus on measurable results.

• Preferred: Experience in sports video computer vision or related fields (such as multi-agent tracking, pose estimation, and crowded scene analysis).

• Knowledge of video tools (e.g., FFmpeg), efficient dataset formats (like WebDataset/shards), or methods for streaming/batching to GPUs.

• MLOps/production experience, encompassing model packaging, continuous integration for training/evaluation, serving (Triton/TorchServe), and monitoring.

🏝️ Benefits

• Competitive Salary and Bonus Plan.

• Comprehensive health insurance coverage.

• Retirement savings plan (401k) with company matching.

• Flexible remote working environment.

• Generous, unlimited time off policy.

• Extensive paid holiday schedule - 13 days total, including the Monday following the Super Bowl.

• Annual performance-based bonus.

• Additional benefits and/or other applicable incentive compensation plans.