Data Scientist

First Horizon Bank•Nashville, TN

3d•Onsite

About The Position

The Data Scientist will play an important role on the Performance Marketing Analytics team with impact across the company. This exciting role provides the opportunity to be a critical part of a growing team that is driving our company’s capabilities into the future. The ideal candidate will blend strategic thinking, strong organizational skills, excellent data science and data management expertise and the ability to develop trusted relationships with a broad range of stakeholders. An understanding of data governance, data risks, data routines, data systems, technology platforms, and data tools will be important attributes for success.

Requirements

7+ years of hands-on data science experience (or equivalent) delivering productionized models and measurable business impact.
Strong Python skills (pandas, numpy, scikit-learn); experience with survival analysis (lifelines, scikit-survival), logistic regression, and tree-based/gradient boosting methods.
Demonstrated MMM experience (e.g., adstock/carryover, saturation, lag modeling; frameworks such as Robyn or LightweightMMM) and translating results into budget recommendations and positive business change.
Experience working with multi-touch attribution models within customer journeys where brand (e.g. awareness, consideration/intent), campaign effectiveness (e.g., lead generation, conversions, etc.) and total investment considerations are leveraged to measure return on marketing investment (ROMI).
CLV modeling expertise (e.g., Pareto/NBD, BG/NBD, Gamma-Gamma, contractual/subscription CLV) and linking CLV to acquisition/retention strategies.
Proven ability to work in scrappy data environments: stitching disparate sources, resolving data quality issues, and iterating quickly with imperfect inputs.
Databricks and big data: Proficiency with SQL, PySpark, MLflow, and Databricks Jobs/Workflows; experience building feature pipelines and monitoring in the platform.
Experimentation: Designing and analyzing A/B and incrementality tests; understanding of causal inference concepts.
Version control and workflow: CI/CD familiarity; reproducible research practices and structured documentation.
Communication and influence: Track record of driving decisions with clear storytelling and stakeholder alignment.
Mentoring experience: Guiding junior associates, establishing coding standards, and elevating team practices.
Experience with cloud data ecosystems (Azure/AWS) and orchestration tools (Airflow, dbt).
Experience with visualization tools (e.g., Plotly, Matplotlib/Seaborn, Power BI/Tableau).

Nice To Haves

Financial services or marketing analytics domain exposure; understanding of client lifecycle in banking or adjacent industries.

Responsibilities

Lead end-to-end development of machine learning and advanced analytics solutions that drive client acquisition, retention, and operational efficiency.
Partner with cross-functional associate’s in Marketing, Digital, Product, Finance, and Data Engineering to translate business goals into measurable data science initiatives.
Build data models in a scrappy environment today, while helping evolve practices and platforms toward a more sophisticated enterprise data structure on Databricks.
Business problem framing: Work with stakeholders to define hypotheses, success metrics, and testable use cases (e.g., acquisition propensities, churn/retention, cross-sell, marketing effectiveness).
Data discovery and wrangling: Source, clean, and integrate messy, siloed datasets; design robust data pipelines and reproducible feature sets.
Modeling and experimentation: Client outcomes: Survival analysis for churn/retention and tenure; logistic regression and tree-based models for acquisition/propensity; CLV modeling. Marketing effectiveness: Build/refresh Marketing Mix Models (MMM) for incrementality, channel contribution, and optimization. Productionalize: Package models with MLflow, implement monitoring/alerting, and manage lifecycle updates.
Measurement and optimization: Produce response curves, elasticities, and scenario planning for budget allocation; quantify short- vs long-term effects where feasible.
Communication: Convert complex analyses into clear narratives, visuals, and recommendations for executive and non-technical audiences; document methods and assumptions.
Mentorship and standards: Coach junior data scientists/analysts through code reviews, pair programming, and project scoping; contribute to best practices, templates, and reusable components.
Governance and risk: Adhere to model governance, validation, fairness, and monitoring standards suitable for a regulated environment; ensure privacy/security controls in data use.