Computer Vision Engineer

SumerSports is hiring a Computer Vision Engineer to build and improve sports video intelligence models, including detection, tracking, pose estimation, event understanding, and multi-view reasoning. This is a CV-first role focused on research and applied modeling, with collaboration on dataset design and productionization. The level (mid vs senior) depends on scope ownership and independence.

• Build and train CV models for sports video: player/ball detection, multi-object tracking, pose/keypoints, event/action recognition, and identity association (re-ID).
• Own the experimentation loop: hypotheses, ablations, error analysis, and measurable improvements.
• Design and maintain evaluation metrics, dataset slices, and failure taxonomies.
• Improve data efficiency through augmentations, sampling strategies, and handling label noise.
• Prototype and iterate on modern architectures such as transformer-based detection/tracking and temporal models.
• Collaborate on dataset and labeling design, formats, schemas, and versioning.
• Help productionize models with packaging, batch/stream inference, and robustness checks.

• Strong applied CV experience with hands-on model development.
• Solid PyTorch skills: training loops, debugging, data pipelines for vision workloads, and DDP basics.