Lead Quantitative Analyst, Computer Vision

Summary

Join a team of cutting-edge researchers in the Phillies Baseball Operations department and help shape the future of baseball by building computer vision models and pipelines. As a Lead Quantitative Analyst, Computer Vision, you will construct predictive metrics and output to improve player evaluation and development. You will leverage cutting-edge computer vision methodologies and techniques to extract meaningful signal from video data for downstream models and analysis. This role requires collaboration with software engineers and other analysts to ensure model output aligns with application needs. You will also contribute to maintaining and growing the computer vision research roadmap and work with the Infrastructure and Machine Learning Engineering team to build robust pipelines. This position offers the unique opportunity to apply your computer vision skillset to a dynamic and innovative environment.

Requirements

• Possess or are pursuing a BS, MS or PhD in Machine Learning, Computer Science, or related or equivalent practical experience
• Demonstrated experience or a public portfolio of applied computer vision work leveraging open-source frameworks such as PyTorch, TensorFlow, Keras, OpenCV, etc
• Willingness to work as part of a team on complex projects
• Proven leadership and self-direction

Responsibilities

• Construct computer vision models using available image and video data to extract signal for use in predictive models and other analyses
• Communicate with software engineers and other quantitative analysts to ensure computer vision model output aligns with the needs of downstream models and applications
• Collaborate with departmental leadership to maintain and grow our computer vision research roadmap
• Work with our Infrastructure and Machine Learning Engineering team to build robust, scalable, and maintainable pipelines for computer vision solutions

Preferred Qualifications

• Familiarity with best practices in machine learning operations (Git, Docker, MLFlow or the equivalent)
• Knowledge of the state of public baseball analytics research