Senior Director, Data Science - NextGen Forecasting - Remote

UnitedHealth Group•Minnetonka, MN

2d•$159,300 - $273,200•Remote

About The Position

Optum Tech is a global leader in health care innovation. Our teams develop cutting-edge solutions that help people live healthier lives and help make the health system work better for everyone. From advanced data analytics and AI to cybersecurity, we use innovative approaches to solve some of health care’s most complex challenges. Your contributions here have the potential to change lives. Ready to build the next breakthrough? Join us to start Caring. Connecting. Growing together. The health solutions marketplace is hungry for new ideas, innovative products and software that drives elevated performance for the business and the customer. The UnitedHealth Group family of businesses is feeding incredible solutions to that marketplace every day by bringing out the best in our software engineering teams. We serve customers across the health system. Not only do we have more of them every day, we also have more technology, greater data resources and far broader expertise than any competitor anywhere. We're out to change the way our businesses and consumers engage with technology. If you're in, you'll be challenged like never before. It's time to join this history making. You’ll enjoy the flexibility to work remotely from anywhere within the U.S. as you take on some tough challenges. For all hires in the Minneapolis or Washington, D.C. area, you will be required to work in the office a minimum of four days per week.

Requirements

Bachelor’s degree in Data Science, Statistics, Mathematics, Computer Science, Engineering, or related field
15+ Years of overall experience in data science, forecasting, predictive, and prescriptive analytics at enterprise scale
Proven experience leading multi layer data science organizations or highly complex analytics programs
Deep expertise in time series forecasting, statistics, machine learning, simulation, optimization, and advanced mathematical techniques
Hands-on experience with forecasting methodologies including ARIMA/SARIMA, exponential smoothing (ETS), state space models, and machine learning based forecasting approaches
Demonstrated expertise in handling non-stationary time series data, seasonality decomposition, trend modeling, temporal feature engineering, and forecasting validation methodologies
Experience applying machine learning techniques (e.g., XGBoost, LightGBM) to forecasting problems including lag-based feature engineering, rolling window statistics, and proper handling of temporal dependencies
Solid understanding of forecasting evaluation metrics and validation approaches including MAE, RMSE, MAPE, rolling window validation, and backtesting frameworks
Demonstrated ability to drive analytically rigorous solutions that influence strategic, operational, and financial decision making
Exposure to Gen AI skill - Large Language Model, RAG

Nice To Haves

Master’s or PhD in Data Science, Statistics, Applied Mathematics, Operations Research, or related discipline
Experience in healthcare, actuarial, or healthcare economics analytics environments
Experience leading high visibility, enterprise scale AI or advanced forecasting programs
Experience using advanced analytics platforms and tools including SQL, Python, R, Hadoop, and large scale data technologies
Experience with advanced forecasting frameworks and libraries such as Prophet, statsmodels, darts, Nixtla, scikit-learn, TensorFlow, or PyTorch
Solid background in forecasting governance, validation, and analytical risk management, and probabilistic forecasting methodologies
Familiarity with sequence-to-sequence forecasting architectures, Transformer models, and modern deep learning approaches for time series forecasting

Responsibilities

Provide enterprise level leadership for advanced forecasting, predictive, and prescriptive analytics supporting strategic, operational, and financial decision making
Lead the design, execution, and scaling of next generation forecasting capabilities leveraging complex, unstructured, and high volume datasets
Apply advanced statistical modeling, machine learning, simulation, optimization, and mathematical techniques to deliver materially earlier insight into emerging trends and risks
Translate complex and ambiguous business questions into analytically rigorous, scalable forecasting solutions delivering measurable enterprise value
Provide strategic direction and accountability for forecasting architecture, modeling strategy, prioritization, validation, and deployment across the Next Gen Forecasting program
Direct multiple layers of management and senior level data science professionals, ensuring strong technical rigor, delivery discipline, and talent development
Establish forecasting standards, governance, and validation frameworks ensuring accuracy, interpretability, scalability, and sustained stakeholder trust
Partner closely with Finance, Actuarial, Healthcare Economics, and Technology leaders to align forecasting roadmaps, manage cross functional dependencies, and embed insights into enterprise decision workflows
Design and operationalize advanced time series forecasting solutions using classical statistical methods (ARIMA/SARIMA, ETS, state space models) as well as modern machine learning and deep learning approaches
Lead development of forecasting frameworks that explicitly account for trend, seasonality, stationarity, autocorrelation, and temporal dependencies across large scale enterprise datasets
Establish enterprise forecasting standards including backtesting methodologies, rolling/expanding window validation, probabilistic forecasting, prediction interval generation, and analytical risk management practices
Drive advanced feature engineering strategies for time series forecasting, including lag features, rolling statistics, calendar effects, Fourier terms, and incorporation of exogenous variables such as events, holidays, and external business drivers
Lead implementation of multi-step forecasting strategies including recursive, direct, and hybrid approaches, leveraging sequence modeling architectures such as LSTM and Transformer-based forecasting models where appropriate
Ensure forecasting solutions appropriately address time-based validation, temporal data leakage prevention, model interpretability, scalability, and sustained stakeholder trust