Senior Data Scientist - Real-Time Esports Predictions

Lead the research, design, and continuous improvement of core predictive models for real-time esports betting. You will be the driving force behind the math, statistical logic, and feature engineering that make models highly accurate and profitable, tackling complex problems in high-frequency data and bridging theoretical research with live product features.

What you'll do

• Lead Model R&D: Design, build, and optimise the machine learning models and statistical frameworks that power real-time odds and betting markets.
• Advanced Feature Engineering: Extract deep predictive signals from raw, high-frequency esports telemetry, turning complex in-game mechanics into structured modelling features.
• Build state-of-the-art models: Focus on model performance and probability calibration. Design rigorous backtesting frameworks to prevent data leakage and evaluate performance against historical market baselines.
• Develop Market Logic: Create the mathematical rules and probabilistic derivations that translate baseline win probabilities into complex derivative markets (handicaps, totals, player props).
• Deploy real-time production systems: Ensure your models are seamlessly translated into production-grade pipelines and microservices.

What you'll need

• 5+ years of professional experience in data science, quantitative research, or statistical modelling.
• Deep, intuitive understanding of probability, statistics, and machine learning theory.
• Expert-level skills in the Python data stack with ability to write clean, production-grade code.
• Proven experience designing complex backtesting environments and defining custom evaluation metrics for unique business problems.

Nice to have

• Deep knowledge of competitive esports (CS2, Dota 2, LoL), the underlying game mechanics, and the competitive meta.
• Experience modeling off streaming data or data that updates continuously over time.
• Understanding of modern MLOps principles and experience with tools like MLFlow, Airflow, etc.