Here's a little bit about me!

I'm an aspiring data scientist and incoming data science intern at Roblox! Previously, I was a data science intern at Gas South and Viasat. Graduating from Georgia Tech this Spring 2020 with a B.S. in Industrial Engineering, I plan to attend graduate school in the Fall for an M.S. in Analytics.

Code. Teach. Dance.

I enjoy performing most data science tasks in Python and visualizations in Seaborn and D3. One of my greatest passions is in teaching, and I've been a TA for Statistics and Regression courses for 2 years, winning the 2019 Outstanding Undergraduate TA award. I also founded and currently lead Seoulstice, a K-Pop dance team that has won at competitions and perform for over 15 events every year!

Featured Projects

Redesigned Event Valuation

For my team's Senior Capstone project, we worked with the Georgia World Congress Center, the third largest convention center in the US, to transform their business model to a data-driven assessment of profitability on a per event basis. We implemented three models to achieve this goal: a rooms-to-event assignment optimization model, a cost prediction model using Gradient Boosting, and a profit margin classification tool. We integrated the three models in a fully interactive web app that is currently implemented. Overall, our project is projected to increase annual profits by $1.1 million (18%). The technical details of our models are contained in the Appendix of our final report.

We were also awarded the Best ISyE project at the Fall 2019 Capstone Design Expo and Best Senior Design project for the course.

Web App Demo

Finalist Presentation

Course Material for ISyE 4031: Regression/Forecasting

I have created Jupyter Notebooks as course material to help students implement and diagnose regression models in Python using StatsModels and Sci-Kit Learn. I also include additional derivation as supplementary material to provide students intuition behind many fundamental concepts, and I review visualization libraries and techniques in Seaborn. I also plan to go through additional concepts related to Machine Learning to supplement course knowledge.

Feel free to check out the collapsible tree diagram below that contains the direct links to the Jupyter Notebooks!

Via-Diet

For a Viasat hackathon, our team created a web app to track employee nutrition intake throughout a work day. By tracking food purchases, we created interactive visualizations to track key macros throughout the day. I used Flask and SQLite for the back-end and HTML/Bootstrap/D3 for the front-end and interactive visualizations.

GitHub Repo

Loan Analysis

Our team analyzed loans from LendingClub to assess characteristics of bad loans and to develop an algorithm for predicting bad loans. We performed extensive EDA, handled imbalanced data, implemented various classifiers, including LightGBM, and assessed fairness and bias concerns in the dataset as well.