Role overview
- Make technical contributions across the automated officiating system, e.g. sensor pipelines, ML data pipelines, training, model development and evaluation pipelines etc.
- Designing, implementing, and deploying state-of-the-art tracking, 3D reconstruction and geometry estimation, scene understanding and visual recognition systems.
- Build and maintain efficient, scalable end-to-end pipelines to manage petabyte-scale multi-modal datasets and model training throughout the entire ML lifecycle.
- Profile, debug and implement tooling to understand bottlenecks and optimize system performance.
- Collaborate with the broader Basketball R&D team on various initiatives, such as sensing research and development, KPI development and measurement, product road mapping, etc.
- Provide technical guidance and mentorship to other engineers on the team.
- Have a strong sense of ownership and be excited to wear many hats.
- Be a guardian of the codebase and push for clean, well-tested and highly extensible code.
- Bachelor’s degree in Computer Science, Electrical Engineering, Math or related field (or equivalent experience).
- Experience working with ML data pipelines and large datasets (TB or PB scale) in a production environment.
- Demonstrated proficiency building and deploying machine learning solutions to production.
- Familiarity with containerization and orchestration frameworks like Kubernetes, Docker.
- Proficiency in Python and prior experience building machine learning data pipelines.
- Proficiency with at least one deep learning framework (Pytorch, TensorFlow, JAX etc).
- Exposure to the entire ML stack, from data pipelines to model inference.
- Excellent problem-solving skills and adaptability in a fast-paced environment.
- Excellent communication and interpersonal skills.
Preferred qualifications
- Proven experience delivering solutions for real-world perception challenges (e.g., AR/VR, autonomous vehicles, robotics, drones).
- Strong C++ programming skills (or another equivalent compiled on-board language), with a history of optimizing and deploying performance-critical systems.
- Familiar with ML training frameworks and prior experience building ML training and evaluation pipelines.
- Experience with production ML systems, including scalable data pipelines, training infrastructure, model evaluation or deployment.
- Familiarity with computer vision libraries, model deployment (TensorRT, ONNX) and GPU acceleration frameworks.
- Strong grasp of low-latency, high-throughput system design, distributed task management systems and scalable model serving & deployment architectures.
- Exposure to CUDA, parallel computing, or high-performance programming on GPUs.
- Passion for basketball and familiarity with officiating rules.
About the company
The National Basketball Association (NBA) is a global sports and media organization with the mission to inspire and connect people everywhere through the power of basketball. Built around five professional sports leagues: the NBA, WNBA, NBA G League, NBA 2K League and Basketball Africa League, the NBA has established a major international presence with games and programming available in 214 countries and territories in 60 languages, and merchandise for sale in more than 200 countries and territories on all seven continents. NBA rosters at the start of the 2024-25 season featured a record-tying 125 international players from a record-tying 43 countries. NBA Digital’s assets include NBA TV, NBA.com, the NBA App and NBA League Pass. The NBA has created one of the largest social media communities in the world, with more than 2.3 billion likes and followers globally across all leagues, team and player platforms. NBA Cares, the NBA’s global social responsibility platform, partners with renowned community-based organizations around the world to address important social issues in the areas of education, inclusion, youth and family development, and health and wellness.