Lead Data Scientist

TEGNA Inc. (NYSE: TGNA) helps people thrive in their local communities by providing the trusted local news and services that matter most. With 64 television stations in 51 U.S. markets, TEGNA reaches more than
100 million people monthly across the web, mobile apps, streaming, and linear television. Together, we are building a sustainable future for local news.

We are seeking a Lead Data Scientist to drive the development and deployment of predictive models for our Sales Automation initiative. You will own the end-to-end data science lifecycle to build machine learning models that predict advertiser probability of conversion to customer, enabling our sales team to prioritize high‑value prospects, optimize outreach strategies, and accelerate revenue growth. Ideal candidates are analytical problem‑solvers who excel at translating business objectives into data‑driven solutions, can work with complex advertiser datasets, and thrive in a fast‑paced, collaborative environment.

• Own the full data science lifecycle for advertiser conversion modeling: problem framing, hypothesis design, feature engineering, model development, validation, deployment, and impact measurement.
• Translate complex analytics and machine learning outputs into consumable business insights by developing Tableau‑or Power BI–like interactive dashboards and generating automated BI reports (Word, PDF, PowerPoint) that support executive decision‑making and revenue strategy.
• Build and optimize predictive models to estimate advertiser probability of conversion using historical sales data, advertiser behavior signals, engagement metrics, and market trends.
• Leverage managed machine learning and predictive modeling capabilities through cloud platforms (e.g., Snowflake Cortex ML, Snowpark ML) to rapidly prototype, rigorously evaluate, and product ionize advertiser conversion probability models, ensuring scalability, reliability, and alignment with sales and revenue business use cases.
• Engineer robust features from multi‑source advertiser datasets including firmographics, engagement history, interaction patterns, and campaign performance; handle data quality issues, missing values, and class imbalance.
• Develop classification models (logistic regression, gradient boosting, neural networks, ensemble methods) with strong emphasis on interpretability and business explainability for sales team adoption.
• Design and implement model validation frameworks: train/test splits, cross‑validation, business‑aligned metrics (precision, recall, AUC, lift), and rigorous backtesting against historical conversion data.
• Establish model governance and monitoring: performance tracking, drift detection, retraining pipelines, fairness assessment, and clear documentation of model assumptions and limitations.
• Create actionable conversion propensity scores and segmentation strategies that enable sales teams to prioritize leads, personalize outreach, and optimize resource allocation.
• Conduct A/B testing and incrementality analysis to measure the business impact of model‑driven sales interventions and continuously improve conversion strategies.
• Translate complex model outputs into clear, executive‑ready narratives and dashboards; deliver recommendations that drive sales strategy and revenue decisions.
• Partner with sales, marketing, and product teams to understand business requirements, prioritize high‑impact opportunities, and design data‑informed roadmaps.
• Mentor teammates on best practices in predictive modeling, statistical rigor, and responsible AI; promote a culture of experimentation and measurable impact.

• Predictive modeling and classification: logistic regression, decision trees, random forests, gradient boosting (XGBoost, Light
  
  GBM), neural networks, and ensemble methods.
• Data Visualization:
  Create clear, compelling data visualizations and dashboards to enable team‑led exploratory analysis and effectively communicate to both technical and non‑technical stakeholders
• Feature engineering: domain‑driven feature creation, feature selection, handling categorical variables, scaling, and dimensionality reduction.
• Statistical methods: hypothesis…