MY_PROFILE.EXE

Welcome to my website!

⋆˚｡⋆୨୧˚✧˖°.‧₊⋅˚✮ Hi, I’m Julia! I’m a Master’s student in Information Technology at Florida State University, currently earning an Information Architecture certificate. I enjoy combining creativity with technology, especially through web development, data science, and exploring the world of AI. I am working toward a career in data science or AI development and love experimenting with new tools and ideas that help me grow these skills. Outside of tech, I enjoy watching MLB and FSU sports, listening to music, thrifting, and finding new recipes to try. ⋆˚｡⋆୨୧˚✧˖°.‧₊⋅˚✮

DOWNLOAD 1 PAGE RESUME DOWNLOAD MASTER RESUME

PROJECTS.SYS

STITCH & SCOUT

Python Gemini Vision AI Streamlit eBay API Scikit-Learn

An end-to-end computer vision solution designed to automate the tedious market research phase of secondhand reselling. This tool leverages Gemini Vision AI to perform multi-label classification on uploaded garment photos, identifying brand, style, and material. The backend then orchestrates real-time data retrieval via the eBay API, using Scikit-Learn for price-point normalization and outlier detection. By consolidating hours of manual search into a single photo upload, the app provides sellers with statistically grounded pricing recommendations instantly.

VIEW ON GITHUB

SPOTIFY REAL-TIME API

Node.js Vercel Functions OAuth 2.0 REST API CORS Management

A full-stack integration that bridges a static frontend with the Spotify Web API. To bypass the limitations of static hosting, I engineered a serverless middleware layer using Vercel Functions. This backend handles the OAuth 2.0 handshake and refresh token rotation securely, keeping private credentials hidden from the client side. The final implementation uses asynchronous JavaScript to fetch live listening data, providing a real-time "Now Playing" experience with custom visualizers and cross-origin resource sharing (CORS) security protocols.

VIEW BELOW

2018 NWSL EXPECTED GOALS (xG) MODEL

R Logistic Regression Sports Analytics Marginal Effects

A predictive modeling project focused on quantifying shot quality in women’s professional soccer. Using Logistic Regression, I built a model to calculate shot conversion probabilities based on spatial variables like shot location, goalkeeper positioning, and defensive pressure. The study evaluates individual player finishing performance against statistical expectations using Brier score evaluation, providing a data-driven lens through which to view NWSL striker efficiency.

VIEW ON GITHUB

NFL WIDE RECEIVER CLUSTERING

R K-Means Clustering Unsupervised Learning Scouting

Redefining player archetypes through unsupervised machine learning. This project utilizes K-Means Clustering to group historical NFL Wide Receivers into five distinct performance profiles based on standardized z-score metrics. By mapping 2024 Draft Prospects onto these established clusters via elbow plots and cluster visualization, the model identifies 'pro-comparisons' for incoming rookies, bridging the gap between historical data and modern scouting.

VIEW ON GITHUB

SEE MORE ON GITHUB!

CURRENTLY_LISTENING.MP4

NOW PLAYING:

Not Listening

Spotify Offline

PAUSE

STOP