Data Scientist

This role sits within Decision Science & Analytics and supports the ongoing development and execution of product shrink and damages forecasting and mitigation efforts. The position focuses on building and monitoring predictive and deterministic models, automating analytics workflows, and running ad hoc investigations to identify root causes.

Key Responsibilities

• Develop dynamic, production-ready, scalable models designed to generate ROI for both Dollar General and its customers.
• Build forecasting and segmentation approaches that may include predictive and time-series forecasts, store segmentations, and geospatial components .
• Create automated and reusable analytics workflows end-to-end, covering development and maintenance of SQL/Python code through final report or dashboard delivery.
• Perform open-ended ad hoc analysis by defining the business question, selecting appropriate analytics techniques, and delivering insights and recommendations to support decisions.
• Share results and actionable insights with stakeholders using PowerPoint or PowerBI , communicating effectively to non-technical audiences.

General Summary

• Lead the ongoing development and execution of product shrink and damages forecasting and mitigation efforts.
• Collaborate with internal business partners, including Asset Protection, Finance, and Merchandising, along with internal IT resources and the broader Decision Science organization.
• Drive daily work that includes developing and monitoring predictive and deterministic models, automating analytics processes, and conducting deep-dive ad hoc analysis to determine root causes.

Required Qualifications

• Strong problem-solving ability using business judgment and robust quantitative analysis.
• Experience building predictive models from scratch and transitioning them into production, including model comparison and selection, parameterization, error handling, and experience working across dev/test/prod environments.
• Ability to apply correct data preparation and feature engineering methods, including outlier identification and removal, principal components analysis (PCA) , and general data structuring.
• Experience writing code to combine, clean, and prepare data for modeling using combinations of SQL , Python , and PySpark (including pandas, numpy, scikit-Learn, matplotlib, and tensor-flow).
• Proficiency with analytical platforms and distributed compute, including Databricks .
• Practical experience ingesting and manipulating large datasets (millions of records).
• Demonstrated capability to explain complex analytics topics clearly to non-technical audiences, including communicating model accuracy and feature importance.
• Experience with code management tools such as GitLab .
• Strong background applying statistical machine learning techniques to predictive modeling and experience with machine learning libraries.
• Minimum 2-5 years hands-on industry (non-academic) experience in Data Science (or an equivalent quantitative role).
• Minimum education : MS in Data Science, Statistics, Economics, Computer Science, Mathematics, or a related applied quantitative field (preference indicated but not fully captured).
• Bachelor’s in a highly quantitative/STEM field may be considered with the right experience.

Preferred Qualifications

• Experience with geospatial analytics.
• Experience in the retail industry.
• Basic to intermediate skills with Tableau and PowerBI .

Technologies

SQL , Python , PySpark , pandas , numpy , scikit-Learn , matplotlib , tensor-flow , Databricks , PowerPoint , PowerBI , Tableau , GitLab , PCA (principal components analysis).

Location and Experience Level

• Location: Goodlettsville, TN (onsite)
• Minimum experience: 2 years