Research Engineer (AI + Sports)

Core Responsibilities

• Design and develop AI systems for real-time video understanding of live sporting events (player detection, action recognition, spatial analysis, etc.)
• Build robust computer vision pipelines that handle challenging real-world footage (lighting, occlusion, multiple camera angles)
• Explore novel architectures and techniques in modern CV to solve sports-specific problems
• Develop AI systems to extract, aggregate, and interpret game data at scale across multiple sports, teams, and seasons
• Create spatial and temporal analytics frameworks that surface actionable insights from video and sensor data
• Build analytics platforms that scale from single games to league-wide deployments
• Translate video understanding and analytics into engaging, intuitive experiences for millions of fans
• Collaborate on product features that leverage AI (real-time highlights, personalized stats, interactive visualizations, etc.)
• Ensure research outputs move through the full product development lifecycle

CORE GOALS

• Publish Your Work: We intend to publish the work coming out of these research projects. Papers will be published in top-tier CV/ML venues and presented at conferences.
• Bridge Academia & Industry: Work directly with Prof. Priya Narasimhan (Carnegie Mellon University) and her research team to translate academic innovations into applied systems. Mentor CMU students, collaborate on research projects, and shape the next generation of sports AI researchers.
• From Research to Product: Own the path from prototype to production. You'll participate in design reviews, handle real-world deployment challenges, and see your ideas impact actual fan experiences at scale.

Core Requirements

• PhD in Computer Vision, Machine Learning, Computer Science, or a closely related field
• Strong publication track record in top-tier venues (CVPR, ICCV, ECCV, NeurIPS, ICML, ICLR, etc.)
• Deep expertise in modern computer vision techniques: neural networks, object detection, semantic/instance segmentation, action recognition, optical flow, pose estimation, or related areas
• Proficiency in ML frameworks (PyTorch, TensorFlow) and modern deep learning practices
• Strong software engineering fundamentals: Python, Java, AWS, SQL, Redshift, version control, testing, CI/CD
• Demonstrated ability to implement complex systems end-to-end
• Background in sports analytics, sports tech, or applied computer vision (industry, research, or both)
• Genuine enthusiasm for sports and AI
• Genuine enthusiasm for going beyond book learning, and to have ideas go into large-scale production