Develop software programs, algorithms, and automated processes that cleanse, integrate, and evaluate large data sets from multiple disparate sources
Manipulate large amounts of data across a diverse set of subject areas, collaborating with other data scientists and data engineers to prepare reliable data pipelines and shared feature sets for various modeling protocols
Build, validate, and maintain AI (Machine Learning / Deep Learning) models, diagnose and optimize performance, and develop statistical models and analysis for ad hoc business-focused questions
Design and execute structured experimentation and testing (e.g., A/B tests, multivariate tests, RL/bandit experiments) to measure incremental impact, refine models, and continuously improve marketing strategies across brands and channels
Communicate meaningful, actionable insights and model outputs to stakeholders, translating them into clear go-to-market recommendations that influence campaign design, offer strategy, contact strategy, and customer experience
Build cross-functional partnerships and consensus with engineering, product management, central marketing, and brand strategy teams to integrate models into marketing campaigns, journeys, and agentic AI workflows at scale
Influence strategy for the Martech Data Science team, including migration and modernization of models on GCP, accelerating agentic AI–driven decisioning, and strengthening a rigorous, test-and-learn culture across campaigns and customer experiences
Advanced proficiency in R, Python, Spark, Hive (or other MR), and common scripting languages for E2E pipeline development. Advanced proficiency using SQL for efficient manipulation of large datasets in on-prem and cloud distributed computing environments, such as Azure and GCP-based environments
Experience with ML and classical predictive techniques such as logistic regression, decision trees, non-linear regressions, ANN/CNN, boosted trees, SVM, TensorFlow, and visualization packages; familiarity with reinforcement learning / bandit methods and uplift / causal modeling is a strong plus, along with a track record of creating business impact with these methods in marketing or personalization contexts
Ability to work both at a detailed level (data wrangling, feature engineering, model diagnostics) as well as summarize findings and extrapolate knowledge to make strong recommendations for change, explicitly linking models and tests to incremental revenue, margin, customer LTV, and contact efficiency
Ability to collaborate with cross-functional teams and influence product and analytics roadmaps, with a demonstrated proficiency in relationship building and in operationalizing models into campaigns, journeys, and orchestration platforms
Ability to assess relatively complex situations and analyze data to make judgments and recommend solutions, balancing short-term performance vs. long-term customer value, and navigating tradeoffs across revenue, profitability, customer experience, and contact pressure