Student Spotlight: Aaron Payne, Data Analyst
Get the full intelligence
Search transcripts, export clips, track mentions, and explore all topics from “Student Spotlight: Aaron Payne, Data Analyst” inside PodZeus.
Aaron Payne, an MBA student at Georgia Tech’s Scheller College of Business and now a senior insights analyst at Chick-fil-A, shares how he leveraged experiential learning to build a predictive model for Confama, a Colombian social services organization. Facing real-world data challenges—including non-English inputs, manual data entry errors, and economic volatility—Aaron and his team moved beyond a basic ARIMA model to create a hybrid ensemble model using CEREMAX (a seasonal ARIMA with exogenous variables) and XGBoost. The key innovation? Incorporating granular economic indicators from Colombia’s Bureau of Labor Statistics, broken down by industry sector, to improve forecast accuracy amid rapid urbanization and political instability. What sets this project apart isn’t just the technical rigor, but the ethical imperative: every forecast directly impacts vulnerable populations. Aaron also reflects on the intense balancing act of working full-time while pursuing an MBA, urging future students to plan meticulously and learn from others’ experiences. His next goal? Bridging business decisions with advanced AI, particularly agentic AI, to drive operational excellence at Chick-fil-A. The episode reveals a powerful truth: data science isn’t just about algorithms—it’s about context, ethics, and human impact. By treating analytics as 'insights in action,' Aaron demonstrates how data can serve not just efficiency, but equity.
Build hybrid models that combine statistical rigor (CEREMAX) with machine learning (XGBoost) to improve forecast accuracy in volatile environments.
Use exogenous variables like industry-specific economic indicators to enhance predictive models, especially in developing economies.
Include historical disruptions like COVID-19 as indicator variables—not outliers—to capture real-world volatility in forecasting.
Prioritize model interpretability for stakeholders; avoid black-box models when business decisions depend on transparency.
Leverage industry expertise as a foundation—'standing on the shoulders of giants'—to guide model development and validation.
…and 3 more takeaways available in PodZeus
Apologies and a Lost Season
The host opens with three apologies: to listeners for irregular episodes, to guest Aaron Payne for a delayed interview, and to the band Run N' Punch for not crediting them properly. A brief interlude sets up the episode as a 'what-if' season on student life in the AI era.
From Tax to Analytics: Aaron's Career Path
Aaron shares his journey from technical tax transformation at Ernst & Young to general analytics at Atrium Hospitality, culminating in his current role as a senior insights analyst at Chick-fil-A, where he serves as an internal consultant for analytics and supply chain.
Experiential Learning at Georgia Tech: The Confama Project
“It really puts it in perspective how those types of forecasts go and actually help real people.”
Data Realities: Language, Errors, and Gaps
The team faced non-English data, manual entry errors, and missing exogenous variables—challenges that required translation, outlier detection, and data interpolation to ensure model reliability.
Model Selection: From ARIMA to Ensemble
“We created a formula using the RMSE to weight the forecast predictions and ultimately were able to greatly reduce the residuals of our forecast.”
“One hitch or one miscalculation can result in people, real people not receiving benefits that are essential to them.”
“We created a formula using the RMSE to weight the forecast predictions and ultimately were able to greatly reduce the residuals of our forecast.”
“It really puts it in perspective how those types of forecasts go and actually help real people.”
Host
Guest
Colombia
place
Aaron Payne
person
Confama
organization
COVID-19
other
Georgia Tech
organization
Chick-fil-A
organization
CEREMAX
other
ARIMA
other
DANE
organization
XGBoost
other
Get the full intelligence
Search transcripts, export clips, track mentions, and explore all topics from “Student Spotlight: Aaron Payne, Data Analyst” inside PodZeus.
Start discovering podcast insights today
Start with a 7-day trial and explore a growing catalog of popular podcasts. No credit card required.
No credit card required • 7-day trial • Cancel anytime