Hello! I am Sarah He :D

second year Data Science student at UCSD

Experience

Data Science & Human Resources Intern | A Round Entertainment

August 2024 - Present

As a Data Science and Human Resources intern, I applied data-driven methods to address organizational challenges and improve efficiency.

In my Data Science role, I collected and cleaned large datasets through web scraping and built machine learning models (Linear Regression, Random Forest, XGBoost) to predict concert revenue based on factors like ticket prices and venue capacity. I also developed interactive Tableau dashboards that enabled stakeholders to explore trends by artist and time period, driving data-informed decision-making.

In my HR role, I automated attendance tracking using Excel, streamlining reporting processes and reducing manual effort. Additionally, I maintained employee records, optimized communication channels, and supported staffing and performance management strategies.

Through these roles, I developed strong technical and analytical skills while delivering impactful solutions tailored to organizational needs.

Data Visualization Staff | The Guardian

October 2023 - Present

I contribute to UCSD's award-winning student newspaper with weekly publications print and online. As a part of the Data Visualization Section, I aggregate and interpret large data sets to produce informative and interactive graphics.

Webmaster | The Guardian

January 2024 - November 2024

As the Webmaster for The UCSD Guardian, I managed and updated the website to ensure fresh, engaging content while maintaining over 80 staff profiles and optimizing site navigation. I also contributed to a major redesign, enhancing user experience and modernizing the site's functionality to better serve its audience.

Projects

Major Power Outage Analysis

Machine Learning, Hypothesis Testing

🔗

I analyzed power outage data to identify factors driving severity, focusing on regional and infrastructural influences. Using exploratory data analysis, feature engineering, and hyperparameter tuning, I developed a logistic regression model predicting severe outages with 85% accuracy while minimizing bias across climate categories. Additionally, I conducted permutation testing to assess missingness dependency, evaluate fairness, and uncover key trends.

EcoMenu

Web Development, Data Analysis, Data Visualization

🔗

I collaborated with two teammates to develop this recipe recommendation platform that suggests users with sustainable recipes based on the carbon footprint of their ingredients. I sourced and cleaned datasets on the carbon footprint of over 350 food ingredients and 2100 recipes. I contributed to the frontend development of the web application and presented visualizations to showcase insights from the recipe and ingredient data.

San Diego Crime Analysis

Data Analysis, Data Visualization

🔗

I cleaned and analyzed data from San Diego Police Department to identify trends that are present in San Diego crimes over the past 4 years. Using Tableau, I created a dashboard to highlight summary statistics and relevant trends, while allowing users to filter through crime types and get information specific to those categories.

Starry Sticker Shop Analysis

Data Analysis, Data Visualization

🔗

Using Python, I analyzed online shop data from the small business I co-founded to assess sales patterns and customer sentiment. I then created interactive visualizations using Python and Tableau to showcase shop performance.

Portfolio Website

Web Development

This website here! I built and designed everything you see on this page from scratch using HTML/CSS and JavaScript.

Dino Diet

Machine Learning, Web Development

🔗

At DataHacks, my team and I implemented a XGBoost classifier that would predict the diet of a dinosaur based on four features with 93% accuracy. We developped a Flask application for users to selected their own dinosaur features along with data visualizations on our webpage to showcase our statistics.

Pasta Detector

Machine Learning, Web Development | February 2024

🔗

At FullyHacks, I worked in a team of 3 to develop an image classification web application to distinguish between different pasta types. We trained a machine learning model to classify pasta images and implemented a frontend web page for users to upload their own photos.

Other

Photography

I'm a photographer for UCSD's student newspaper The Guardian staff. My passion for photography and student journalism stems from my former participation as Photography Editor of Temple City High School's Templar Yearbook.