Hybrid Lead Data Scientist – Forecasting

at CVS Health

• Job title: Lead Data Scientist – Forecasting
• Job type: Full Time
• Experience level: Senior
• Salary: $142,140 - $284,280 per year
• Degree requirement: Bachelor's Degree
• Tech skills: AWS, Azure, Cloud, Docker, Google Cloud Platform, Kubernetes, Python, Spark, SQL
• Location requirements: Hybrid, New York City, United States

About the role

• Lead Data Scientist advancing forecasting models for pricing and promotions at CVS Health. Collaborate with cross-functional teams to integrate diverse datasets and deliver business impact.

Responsibilities

• Build, optimize, and deploy scalable forecasting models that support pricing, promotions, and assortment strategies across multiple product categories
• Apply advanced statistical, machine learning, and deep learning methods (e.g., ARIMA, Prophet, gradient boosting, LSTMs, hybrid ensembles) for forecasting at SKU, category, and chain levels
• Implement robust MLOps practices for model deployment, monitoring, and retraining using cloud platforms (Azure, GCP, AWS)
• Integrate multiple internal and external data sources (e.g., coupon redemption, merchandising, competitive, and macroeconomic data) into forecasting pipelines
• Collaborate with data engineering to ensure scalable, high-quality data pipelines
• Partner with business stakeholders in pricing, promotions, and assortment to design and validate forecast-driven decision workflows
• Coach and mentor junior data scientists, sharing best practices in forecasting, MLOps, and applied analytics
• Monitor forecast accuracy, perform backtesting, and refine models to reduce error rates and improve stability
• Develop frameworks for scenario planning and simulation to measure business impact of promotions, pricing strategies, and assortment changes

Requirements

• 7+ years of experience in data science, forecasting, or applied predictive modeling
• 4+ years of experience building and deploying time-series forecasting models using methods such as ARIMA, Prophet, gradient boosting, LSTMs, or hybrid ensembles
• 4+ years of experience with Python and SQL for large-scale data processing
• 3+ years of experience with MLOps tools and practices (e.g., GitHub/GitLab, Docker, Kubernetes, Kubeflow, CI/CD pipelines)
• 3+ years of experience using cloud platforms (Azure, AWS, or GCP) and distributed computing frameworks (e.g., Databricks, Spark)
• Proven track record of deploying at least 2 production forecasting models that delivered ≥10% improvement in accuracy (e.g., reduction in MAPE, WMAPE, or sMAPE
• Experience working with cross-functional teams (engineering, merchandising, pricing, assortment) to deliver at least 2+ enterprise-level data-driven solutions from design to production

Benefits

• Affordable medical plan options
• 401(k) plan (including matching company contributions)
• Employee stock purchase plan
• No-cost programs for all colleagues including wellness screenings, tobacco cessation and weight management programs, confidential counseling and financial coaching
• Benefit solutions that address the different needs and preferences of our colleagues including paid time off, flexible work schedules, family leave, dependent care resources, colleague assistance programs, tuition assistance, retiree medical access and many other benefits depending on eligibility