Job Description

PROBLEMS YOU WILL HELP SOLVE

• Personalisation & recommendations — Recommend relevant challenges, rewards, offers and in-game experiences using player behaviour, context and preferences; evolve from broad segments toward individual-level personalisation.
• Potential payer & value prediction — Identify players likely to make a first purchase, estimate future player value (LTV) and support thoughtful decisions about when and how to engage different player groups.
• Churn prediction & retention — Detect early signs of disengagement, understand why players leave and develop models that help target meaningful retention interventions.
• Player economy & monetisation — Find opportunities across virtual currency, game modes, in-app purchases and ads while balancing monetisation with healthy gameplay and player experience.
• Experimentation & incremental impact — Use A/B tests and causal approaches to evaluate whether model-driven interventions actually improve outcomes, not just predictions.

WHAT YOU WILL OWN:

• End-to-end applied ML — Frame product questions as measurable ML problems; develop features, train and evaluate models, and help operationalize scoring or recommendations.
• Rigorous measurement — Choose suitable offline and online metrics; address rare outcomes, data leakage, bias, calibration and model drift; design trustworthy experiments and guardrails.
• Cross-functional delivery — Partner closely with Product, Game Design, LiveOps, Analytics and Engineering to turn models into decisions, features and measurable results.
• Clear communication — Explain trade-offs, insights and recommendations to both technical and non-technical stakeholders; help shape the data science roadmap.

MOONFROG LABS · SENIOR DATA SCIENTIST

WHAT WE ARE LOOKING FOR

• Experience — 4+ years in data science or applied machine learning, with evidence of taking models beyond notebooks into real product or business decisions.
• Technical strength — Strong Python and SQL skills; practical experience with feature engineering, supervised learning, model evaluation and libraries such as scikit-learn, XGBoost/LightGBM, PyTorch or TensorFlow.
• Statistical judgment — Solid grounding in probability, statistics, experimentation and interpreting model performance for imbalanced classification, ranking or regression problems.
• Execution mindset — Comfort working with large behavioral/event datasets, ambiguous problems, iteration, version control and cloud environments such as AWS, Azure or GCP.
• Product curiosity — An ability to connect technical results to player behavior, user experience and measurable business impact.

GREAT TO HAVE

• Domain experience — Free-to-play/mobile gaming, consumer apps, recommendations, churn prevention, payer propensity, LTV forecasting or lifecycle marketing.
• Advanced methods — Recommender systems, uplift modeling, contextual bandits, reinforcement learning, causal inference, deep learning or production ML/MLOps.
• Emerging technologies — Interest in applying LLMs or agentic AI to meaningful analytics or player-experience problems.