Soccer Data Scientist II — Recruitment Analytics

Openkrill; United States; Remote; USD 110,000 - 170,000; Full time

Job summary

Teamworks is hiring a Data Scientist II on our Soccer team to develop statistical and machine learning models used by partner clubs for player recruitment. The role emphasizes models built from player tracking data and event data to identify undervalued players and characterize the skills that make them effective.

You will own the model lifecycle from design through deployment and improvement, collaborating with engineering to keep models accurate and reliable.

Qualifications

• 3+ years of professional or research experience writing code in R or Python for data science, statistics, or machine learning work
• Comfort with SQL for building and querying structured data
• Strong data science fundamentals: you understand how models work, what they need from the data, and how to set data up to support them
• Experience working with sports tracking data, including understanding the kinds of models typically built and the data challenges that come with them
• Strong communication skills — you'll be explaining modeling decisions directly to partner clubs, not just internal teammates

Responsibilities

• Develop statistical or ML models using Python or R for player recruitment
• Own the model lifecycle from design and validation through deployment and improvement
• Work with sports tracking and event data to identify undervalued players and characterize useful skills
• Collaborate with engineering to deploy models in production and monitor performance
• Explain modeling decisions and results clearly to technical teammates and professional clubs

Skills

Python

R

Machine learning

Statistical modeling

SQL

Data visualization

Code quality

Tools

AWS

Google Cloud

Job description

Teamworks is hiring a Data Scientist II on our Soccer team to develop statistical and machine learning models used by partner clubs for player recruitment. The role emphasizes models built from player tracking data and event data to identify undervalued players and characterize the skills that make them effective.

You will own the model lifecycle from design through deployment and improvement, collaborating with engineering to keep models accurate and reliable.