Currently an Extern @Hydroficient·🟢 Open to work
I’m a Computer Science and Economics graduate passionate about using data to solve real-world problems. My work spans data analytics, machine learning, predictive modeling, and business strategy.
Explore my work in data analytics and cybersecurity, from building models that uncover patterns in healthcare, forecasting business performance, to focusing on IoT defense strategies.
.jpg)
Hydroficient · Jul 2026 · ✅ Verified by Extern
Build, hack, and defend real IoT infrastructure, then hand your security playbook to an actual company. You’ll launch realistic cyberattacks, from spoofing and replay attacks to unauthorized shutoffs, and implement defenses like TLS and mTLS.
This project applies advanced machine learning methods to understand the drivers of COVID-19 vaccination uptake using a large, nationally representative dataset. The study emphasizes both predictive accuracy and interpretability, contributing to public-health decision-making and academic research.
I’m a Computer Science and Economics graduate passionate about using data to solve real-world problems. My work spans data analytics, machine learning, predictive modeling, and business strategy.
I majored in Computer Science and Economics, with a focus on Data Analytics. I thrive on tackling complex challenges in technology and security, particularly in the realm of IoT infrastructure. A unique ability to connect technical analysis to business and marketing strategy
Externships
Hydroficient IoT Cyber Defense Externship
Hydroficient
Experience
Front Desk Coordinator
Delc Medical Center · Aug 2022 - Aug 2025
Education
Austin College
BS.c Computer Science and B.A Economics · Class of 2026
Skills
✅ Verified by Extern
Build, hack, and defend real IoT infrastructure, then hand your security playbook to an actual company. You’ll launch realistic cyberattacks, from spoofing and replay attacks to unauthorized shutoffs, and implement defenses like TLS and mTLS.
The work recreated a smart traffic-light and water-sensor IoT environment, mapped assets with confidentiality, integrity, and availability ratings, and performed a STRIDE analysis per component. It produced a Python sensor-log generator, an intercepted MQTT trace from an unsecured pipeline, TLS experiment metrics, a Device Provisioning Policy for mTLS, replay-attack test results, and a real-time
I delivered a prioritized STRIDE threat model, exported structured sensor logs from a Python generator, an intercepted MQTT message trace, measured TLS metrics (latency and throughput), a Device Provisioning Policy for mTLS, a replay-defense report, and a dashboard-backed experiment showing all tested attacks blocked.
This project applies advanced machine learning methods to understand the drivers of COVID-19 vaccination uptake using a large, nationally representative dataset. The study emphasizes both predictive accuracy and interpretability, contributing to public-health decision-making and academic research.
This project applies advanced machine learning methods to understand the drivers of COVID-19 vaccination uptake using a large, nationally representative dataset. The study emphasizes both predictive accuracy and interpretability, contributing to public-health decision-making and academic research.
Built and evaluated nine classification models, including Random Forest, XGBoost, SVM, LASSO, and logistic regression, using train/test splits and 10-fold cross-validation. Compared accuracy, precision, recall, F1 and AUC, then used SHAP to interpret predictions and analyze subgroup differences.
Random Forest was the strongest model, achieving a 0.944 AUC versus 0.812 for logistic regression. SHAP analysis identified prior routine vaccination, education, and income-to-poverty ratio as the most consistent predictors, while revealing meaningful differences across regions and demographic groups.
Kalibrate
Built a predictive sales model to determine which of five potential Pizza Planet locations offered the strongest expansion opportunity. Analyzed demographic, geographic, competitive, and site-level factors to identify key sales drivers and translate quantitative findings into actionable real estate
Built a predictive sales model to determine which of five potential Pizza Planet locations offered the strongest expansion opportunity. Analyzed demographic, geographic, competitive, and site-level factors to identify key sales drivers and translate quantitative findings into actionable real estate recommendations.
Cleaned and prepared location data, selected features using correlation analysis and economic intuition, and tested for multicollinearity using VIF. Built and compared linear, nonlinear/interaction, and LASSO regression models, then evaluated performance on holdout data using R², RMSE, and MAE.
The final model achieved a 0.563 R² on holdout data and was used to forecast sales across five proposed locations. Store 9002 ranked highest with approximately $1.67M in predicted sales, followed by Store 9001 at $1.49M, providing a data-backed recommendation for expansion.