I'm a Data Science student at UNC Charlotte (May 2026 Graduate) with real-world experience applying analytics and ML to healthcare research, biopharma strategy, and financial operations. I aim to innovate to help others and spark individual curiosity — from predicting Parkinson's disease, exploring planets beyond our solar system, to tracking all 3 languages I'm actively learning!
I grew up curious — always wanting to understand why things work the way they do. Data Science gave me a framework to satisfy that curiosity: ask hard questions, gather the right evidence, and let the patterns speak.
At Avalyn Pharma, I spent nearly a year interning in Corporate Development and 3 months in Finance — building CRM systems, KPI dashboards, and financial models that were presented directly to the CFO and G&A team. That experience shaped how I think about data work: it's only valuable if it communicates clearly to real decision-makers.
My research has taken me into healthcare analytics, and two of my projects were deeply personal. When my uncle was diagnosed with Parkinson’s disease, I translated that experience into two independent initiatives: an industry research project using SocialBit software under the Chair of Computer Science at UNCC, and a separate machine learning study leveraging clinical voice data to explore early Parkinson’s detection. That's the kind of work that reminds me why data science matters beyond the model metrics.
Outside of data, I've been teaching martial arts for 11 years. It's taught me how to communicate complex ideas, understand different perspectives, and remain confident during difficult situations.
Real problems. Real data. Real results.
Applied machine learning to a clinical dataset of 23,000+ patient records to identify Parkinson's disease from acoustic voice measurements. Features like jitter, shimmer, and pitch irregularity — invisible to the ear, but detectable through signal processing.
Performed EDA, correlation analysis, and dimensionality reduction (PCA) to distill 20+ acoustic variables into the features most predictive of disease status. Visualized results clearly for non-technical stakeholders. Trained and evaluated classification models, addressing class imbalance (4,290 healthy vs. 19,551 Parkinson's records).
Used K-Means and Hierarchical clustering on a 4,855-planet Kaggle dataset to discover whether exoplanets naturally group into known planetary categories — rocky planets, gas giants, Hot Jupiters — using only physical and orbital measurements.
Built custom data cleaning pipelines to handle messy, mixed-format scientific measurements. Applied PCA for 2D cluster visualization. Validated groupings using hierarchical dendrograms. The 3 clusters mapped cleanly to small cool planets, temperate gas giants, and close-orbiting Hot Jupiters.
An active industry research collaboration with SocialBit. Using wearable sensor data — heart rate, movement variance, AI-predicted activity states — to assess whether social interactions in a Parkinson's patient are continuous or fragmented throughout the day.
Active data collection phase from a patient with Parkinson's Disease. Establishing benchmarks from a healthy control baseline to enable comparison against the Parkinson's patient. Results and methodology write-up coming upon study completion.
Applying data skills in real organizations.
Operated across two high-visibility functions with direct exposure to C-suite decision-making for a clinical-stage biopharma company.
11 years of teaching martial arts has made me a better communicator than any course could. I know how to explain complex ideas to someone who's never heard them before — whether that's a six-year-old learning a kick or a CFO reviewing a dashboard. I show up curious, clear, and ready to listen. That's rare in data science. I'm working to keep it that way.
Reach out about roles, projects, or just to say hi.