Lead Scientist, Forecasting

• Lead the design, development, and deployment of forecasting and pricing models using a blend of classical time series, deep learning and Bayesian statistical techniques
• Develop hierarchical forecasting frameworks, including multi-level Bayesian models, that scale across thousands of hotel properties
• Build uncertainty quantification frameworks to increase trust and robustness in forecasts
• Guide model architecture choices—balancing complexity, interpretability, and operational feasibility
• Collaborate closely with engineering to deploy and monitor models in production (e.g., using AWS SageMaker)
• Translate model outputs into actionable insights in partnership with product and business stakeholders
• Define and execute model performance measurement strategies, including causal inference and uplift modeling
• Present findings, experimental results, and strategic recommendations to senior leadership

• MS or PhD in Statistics, Econometrics, Computer Science, Operations Research, or a related quantitative field
• 5+ years of experience delivering impactful data science solutions in production environments
• Expertise in time series forecasting, including classical methods (e.g., ARIMA, Exponential Smoothing, State Space Models) and deep learning (e.g., RNNs, Temporal Fusion Transformers)
• Practical experience with Bayesian modeling, including hierarchical models and probabilistic programming (e.g., PyMC3, Stan)
• Proficiency with ML/DL frameworks (e.g., PyTorch, TensorFlow, scikit-learn, DARTS) and programming languages (Python, R, SQL)
• Familiarity with cloud platforms and MLOps tools (e.g., AWS SageMaker, MLflow) for scalable model development and deployment
• Strong communication and presentation skills, capable of conveying complex analytical concepts to non-technical stakeholders
• Experience designing model evaluation and impact measurement frameworks, including causal inference

Prior experience in the hospitality, travel, or revenue management domain is highly desirable