I am an established professional presenting selected projects and skills that reflect my hands-on experience.
A curated set of projects, case notes, and presentations that show the work I led, tools I used, and the results I delivered.
Developed hypotheses linking volatility, rates, inflation, employment, and sector behavior to future market stress. Engineered 18 features using SQL CTEs, window functions, rolling metrics, and publication-lag controls. Backtested 986 out-of-sample observations, achieving 0.625 ROC-AUC and documenting look-ahead bias controls.
Automated commodity contract valuation with Python, reducing processing time from 45 minutes to under five. Modeled 50,000+ borrower records, improving default-prediction accuracy 20% over baseline.
I am an established professional presenting selected projects and skills that reflect my hands-on experience.
I am Briana Love Walker Whylie. I am firmly established in my career and focus on hands-on work that shows what I can do. I write and present my experience through a curated portfolio that highlights projects and skills.
Experience
Data Scientist Apprentice, Finance Department
Seentio Financial Software · March 2026 - Present
Data Science Summer Intern
Kvadrat · Summer 2026
Union Ironworker Apprentice
Local 580 · June 2021 - April 2025
Accounts Payable, Finance Department
Universal Music Group · March 2019 - March 2020
Education
Pace University, Seidenberg School - New York, NY
Master of Science in Data Science | Concentration: Quantitative Research and Data Science | GPA: 4.0
Berkeley College - New York, NY
Bachelor of Business Administration in International Business | Concentration: Finance | GPA: 3.6 · Class of 2018
Harvard Online
Data Science with Python Coursework · Class of 2022
Skills
Developed hypotheses linking volatility, rates, inflation, employment, and sector behavior to future market stress. Engineered 18 features using SQL CTEs, window functions, rolling metrics, and publication-lag controls. Backtested 986 out-of-sample observations, achieving 0.625 ROC-AUC and documenting look-ahead bias controls.
Automated commodity contract valuation with Python, reducing processing time from 45 minutes to under five. Modeled 50,000+ borrower records, improving default-prediction accuracy 20% over baseline.