Quantitative Data Scientist

Job Description

Job Title: Quantitative Data Scientist (Modeling and Feature Engineering)

\n

Location: New York, NY (In-person)

\n

Experience: 2+ years

\n

Compensation: $80,000-$135,000 base salary + equity

\n

\n

Company Overview: ShotQuality is a leader in AI-driven sports data extraction, combining advanced computer vision and data science to transform video feeds into granular player tracking data and predictive insights. Originally built around basketball, our technology now scales across football, baseball, tennis, and more, powering multi-sport analytics at the enterprise level. Our proprietary models are trusted by industry leaders, enabling sportsbooks, prediction markets, trading firms, and media companies to make confident, data-driven decisions. With 20,000+ games processed across 15+ leagues, ShotQuality turns raw video into actionable intelligence. As we expand our AI and computer vision capabilities, we continue to push the boundaries of real-time sports analytics, delivering the infrastructure that powers the future of data in sports.We are a small team in New York where the work you ship shows up in live products within days, not months.

\n

\n

Job Overview: We are hiring a Quantitative Data Scientist to build the features and predictive models behind ShotQuality's real-time sports products. You will turn raw event, tracking, and market data into signals, test whether they improve on market prices, and get the ones that work into production. This role sits between our data science team and our live products, so you will see your models used in real time by market-facing clients during live games. The ideal candidate comes from a sports and trading background and is energized by fast-moving markets.

\n

\n

Responsibilities:

\n

\n
• Feature Engineering: Design, build, and maintain features from play-by-play, player tracking, and pricing data that improve predictive accuracy.
\n
• Predictive Modeling: Develop and refine probability models for in-game and pregame outcomes, with a focus on calibration and real-world performance.
\n
• Market Analysis: Study how prices move around in-game events and identify where our models add information beyond current market prices.
\n
• Backtesting and Evaluation: Build rigorous backtests that measure whether a model or signal holds up out of sample, against closing lines and live market prices.
\n
• Research to Production: Partner with engineering to move models from notebook to live, low-latency pipelines, and monitor them once they are running.
\n
• Performance Analysis: Track model performance over time, diagnose drift or degradation, and report findings clearly to leadership.
\n
• Client Support: Handle requests from sophisticated, market-facing clients with fast, well-reasoned analysis.
\n

\n

\n

The Role is for Candidates Who: 

\n

\n
• Have a background in trading, market making, or quantitative research and get excited about finding an edge in live markets
\n
• Have built predictive models that were evaluated on real outcomes and real prices, not just a held-out test set
\n
• Enjoy the feature engineering grind and can explain which features mattered and why
\n
• Think in probabilities and expected value, and care about calibration, not just accuracy
\n
• Move fast, ship quickly, and iterate based on what the data shows
\n
• Adapt quickly when priorities shift and are comfortable working without a fixed playbook
\n
• Are comfortable with unstructured hours that follow the sports calendar, including live games on nights and weekends
\n

\n

\n

The Role is NOT for Candidates Who:

\n

\n
• Have only coursework or Kaggle-style modeling experience
\n
• Aren't genuinely interested in prediction markets and the sports betting space
\n
• Prefer pure research with no path to production
\n
• Need a rigid structure or a predictable 9-to-5 schedule
\n
• Are looking for a fully remote role
\n

\n

\n

Requirements:

\n

\n
• 2+ years of professional experience in quantitative research, trading, or data science, ideally at a trading firm, sportsbook, or prediction market
\n
• Strong foundation in statistics and probability, including model calibration and evaluation
\n
• Proficiency in Python (pandas, TensorFlow, NumPy, scikit-learn or similar) and SQL
\n
• Hands-on experience with gradient boosting, regression, and time-series or sequential modeling
\n
• Experience designing backtests and avoiding leakage and overfitting
\n
• Ability to communicate results clearly to technical and non-technical stakeholders
\n
• Bachelor's degree in Data Science, Statistics, Mathematics, Computer Science, Economics, or a related quantitative field
\n
• Willingness to be available during some nights and weekends of live games (as necessary)
\n

\n

\n

Nice to Have:

\n

\n
• Experience with pricing, market making, or market microstructure in sports betting or prediction markets
\n
• A track record of building and running your own models in live markets
\n
• Experience with sports data, player tracking data, or play-by-play data
\n
• Experience deploying models to real-time or low-latency systems
\n
• Exposure to AWS or GCP, Docker, and workflow orchestration tools
\n

\n

\n

Benefits:

\n

\n
• Health, Vision, Dental, and Life Insurance: Comprehensive plans for all full-time employees that are HSA-eligible
\n
• Unlimited PTO: Take time off when you need it while owning your responsibilities
\n
• 401(k) benefits
\n
• WFH and Mental Health + Wellness stipends to support your home office setup and wellness
\n
• Flexible Hours: Flexibility around in-office days, as long as deliverables are achieved
\n
• Learning and Development: Access to training, courses, and conferences to grow your skills
\n
• Live Game Tickets: Tickets to attend live sports games during the year
\n
• Subscription to Claude/Codex/Gemini Accounts
\n

\n

\n

Application Process:

\n

\n
1. Assessment: Complete a take-home modeling and feature engineering exercise built on real game and market data.
\n
2. Interview: Walk our data science team through your approach and results.
\n
3. Final Round: Meet leadership in person at our New York office.
\n