Data Scientist

Company Description

At Marquee, we're building the AI platform that professional sports clubs use to scout, recruit, and analyze players. We combine deep learning with real match data to give clubs intelligence they can act on, from player chemistry and transfer fit to AI-generated scouting reports and squad optimization. Built by sports professionals for sports professionals, our platform is live with professional football clubs. We're expanding into basketball and additional sports, and we're backed by AnD Ventures and angels including the founders of Wix and the CEO of 365Scores.

Role Description

As a Data Scientist at Marquee, you'll work directly with the co-founder and CDO as the first hire on the data science team. This is a full-stack, high-ownership role. You'll build models, ship them to production, and see them drive real decisions at real clubs.

Responsibilities

• Design, train, and improve machine learning models across a multi-model deep learning stack
• Build and own data pipelines end-to-end, from raw data ingestion to production-ready features
• Develop player intelligence and scoring systems used by professional clubs
• Design and improve AI-generated reports
• Research and implement new methods from the latest sports analytics and ML literature
• Collaborate closely with product and engineering to translate scouting and analytics needs into ML solutions
• Validate models rigorously against labeled datasets and real-world outcomes

Requirements

• Strong programming skills in Python and proficiency with PyTorch, pandas, NumPy
• Experience training and deploying ML models in production environments
• Solid understanding of deep learning fundamentals
• Comfort with data engineering: ETL pipelines, SQL, working with large and messy real-world datasets
• Ability to own projects independently from problem formulation to production deployment
• Experience building LLMs and AI agents
• Experience with cloud infrastructure (GCP preferred)
• Strong analytical thinking and attention to detail

Nice to have

• Experience with graph neural networks, transformers, or foundation models
• Background in causal inference, counterfactual reasoning, or Bayesian modeling
• Experience with multi-objective optimization or decision intelligence systems
• Background in recommendation systems, similarity search, or embedding-based retrieval
• Football or sports domain knowledge, understanding of scouting, player evaluation, or sports analytics (Big advantage)
• 1000+ hours in Football Manager (we won't judge, we'll appreciate)