Computer Vision Engineer

Computer Vision Engineer

FieldCoachAI is a Sports Technology platform transforming how sports are analyzed, graded, and coached. We are moving beyond manual film study by leveraging advanced artificial intelligence and computer vision to deliver automated film breakdown and deep tactical insights. Our engineering team builds tracking and performance grading systems that dissect elite athletic movements, micro-mechanics, alignments—turning raw video data into actionable, professional grade intelligence.

The Role

As a Computer Vision Engineer, you will engineer the core architecture that interprets athletic film. You will design, train, and deploy production-grade models capable of precise object detection, human pose estimation, multi-object tracking, and complex spatial-temporal action recognition. Your work will directly power our automated analysis engine, calculating pinpoint metrics like break angles, acceleration bursts, and separation distance.

Key Responsibilities

• Model Architecture & Training: Design, train, and optimize state-of-the-art deep learning architectures (e.g., YOLO, Roboflow, ViTs, CNNs) for object detection, multi-player tracking, and precise 3D human pose estimation.
• Pipeline Development: Build scalable, end-to-end video processing pipelines capable of handling high-volume, multi-angle game and practice footage.
• Biomechanical Metrics Extraction: Develop geometric and spatial algorithms to extract granular athletic metrics, including joint angles, route break angles, velocity, and release mechanics.
• Performance Optimization: Optimize models for efficient, high-throughput inference (utilizing TensorRT, ONNX, or custom quantization) to ensure rapid video processing turnaround.
• Data Pipeline Ownership: Define dataset requirements, manage data annotation strategy and pipelines, and implement robust data augmentation techniques to handle varying lighting, camera angles, and jersey types.
• Cross-Functional Integration: Collaborate closely with the platform engineering team to integrate model outputs seamlessly into our frontend UI and RAG-driven analytical architectures.

Qualifications & Skills Required (Experience)

• Industry Track Record: 2+ years of professional engineering experience deploying deep learning and computer vision models into scalable production environments.
• Core CV Frameworks: Mastery of Python and deep learning frameworks such as PyTorch or TensorFlow, along with foundational libraries like OpenCV, SciPy, and NumPy. Indeed
• Video Analytics Expertise: Strong hands-on experience with object tracking (e.g., DeepSORT, ByteTrack) and temporal action segmentation in video sequences.
• Mathematical Foundation: Strong grasp of linear algebra, multi-view geometry, vector calculus, and statistical data analysis. OpenCV
• Production & Deployment: Experience with model containerization (Docker), cloud infrastructure (AWS or GCP), and writing clean, maintainable, production-ready code.

Nice-to-Haves

• Experience with YOLOv8 Object Detection, Roboflow.
• Familiarity with edge deployment, custom GPU acceleration, or optimizing pipelines for web-app delivery.
• An interest or background in sports analytics, biomechanics, or computer vision in general.

Work Location: Remote (Must be in U.S)

Pay: $1.00 - $3,000.00 per month

Work Location: Remote