Principal Data Scientist- Decisioning

Citizens Financial Group•Columbus, OH

3d•$149,000 - $194,000•Hybrid

About The Position

The Enterprise Data & Analytics organization is seeking a Principal Data Scientist to design, build, and improve quantitative models that strengthen customer relationships, improve decision quality, and drive measurable business outcomes. The role leads statistical, machine learning, and causal inference solutions from problem definition through deployment, monitoring, validation, and continuous refinement. It uses experimentation, counterfactual analysis, quasi-experimental techniques, and observational data to evaluate targeting and decision performance and improve customer and business outcomes. The Principal Data Scientist develops methods to estimate incremental impact, treatment effects, and heterogeneous customer response, translating ambiguous business objectives into defensible analytical problems, measurable outcomes, requirements, and implementation plans. The role partners with data science, marketing, product, finance, risk, data, and technology teams to integrate models into governed business processes and production decision systems. It communicates analytical findings and recommendations clearly to different audiences, documents assumptions, tradeoffs, limitations, and controls, and improves analytical systems using experimental results, operational outcomes, and observed customer behavior. This is a hybrid role with a schedule of four days per week onsite and one day remote, Monday through Friday, for 40 hours per week. The salary range is $149,000–$194,000 per year, with an opportunity to earn an annual discretionary bonus. The employer offers medical, dental, and vision coverage, retirement benefits, maternity and paternity leave, flexible work arrangements, education reimbursement, wellness programs, and paid time off exceeding applicable mandatory paid sick or paid time-away policies.

Requirements

Master’s degree in statistics, economics, engineering, computer science, or a related quantitative field.
10+ years of experience applying advanced analytics and delivering models through the full development lifecycle, including deployment, monitoring, governance, and ongoing improvement.
Deep knowledge of statistical modeling, causal inference, and experimental design, with demonstrated ability to improve decision quality and targeting through estimation and counterfactual reasoning.
Experience estimating heterogeneous treatment effects and incremental value using randomized experiments, causal machine learning, quasi-experimental techniques, uplift modeling, and contextual bandits.
Strong Python skills, including 5+ years of applied experience.
Strong SQL skills, including 5+ years of applied experience.
Experience building reproducible analytical pipelines and working efficiently with large, complex datasets.
Experience with cloud computing platforms and ML-related services.
Ability to balance analytical rigor, interpretability, implementation complexity, and business usefulness under evolving priorities.
Strong stakeholder-management and communication skills, including ability to explain complex analytical, experimental, and causal concepts clearly.

Nice To Haves

PhD in statistics, economics, engineering, computer science, or a related quantitative field.
Experience with AWS and ML-related services such as Amazon SageMaker.
Financial services experience and familiarity with customer-level behavior, profitability, personalization, and relationship analytics.

Responsibilities

Lead development of statistical, machine learning, and causal inference solutions from problem definition through deployment, monitoring, validation, and continuous refinement.
Evaluate targeting and decision performance using experimentation, counterfactual analysis, quasi-experimental techniques, and observational data.
Develop methods to estimate incremental impact, treatment effects, and heterogeneous customer response.
Translate ambiguous business objectives into defensible analytical problems, measurable outcomes, analytical requirements, and practical implementation plans.
Partner with data science, marketing, product, finance, risk, data, and technology teams to integrate models into governed business processes and production decision systems.
Synthesize analytical findings into clear stories and recommended actions for different audiences.
Communicate assumptions, tradeoffs, limitations, and recommendations clearly, maintaining appropriate documentation, controls, and ownership.
Continuously improve analytical systems using experimental results, operational outcomes, and observed customer behavior to refine models and targeting strategies.
Design, execute, and evaluate experiments, including randomization strategy, holdouts, guardrails, outcome measurement, and causal interpretation.