Master of Computer Science
What do I do?
Graduated with a Bachelors of Science in Data Science and Bachelors of Liberal Arts in Economics, with a minor in statistics at Colorado State University.
Master of Computer Science at the Illinois Institute of Technology
What I have learned so far:
Increased fleet safety and operational efficiency by developing a real-time Streamlit dashboard to identify camera failures (SD, Video Loss). Engineered an automated ETL pipeline using Python (Pandas) and MariaDB to process historical MQTT data, securing sensitive multi-client data with role-based authentication.
Maintained pricing grids for 150+ products, performed pricing analysis, and created reports. Maintained volume forecasts and Power BI dashboards. Executed special projects in financial analysis involving promotion "take-rates" and created presentation decks for senior leadership.
Queried, cleaned, and modeled data using Python (Pandas). Trained a machine learning model using Scikit-learn to predict students more likely to enroll at CSU given the data from their initial application. Deployed the model and presented the results in an interactive dashboard (Streamlit/Seaborn).
Engineered a privacy-preserving RL framework for ICU sepsis simulation using PyTorch and Chameleon Cloud. Implemented Proximal Policy Optimization (PPO) with Federated Averaging and Laplace noise, achieving an 11% performance improvement over baseline models while significantly reducing client identifiability.
In this project, I designed and implemented a database to store Major League Soccer Data. For the design process I used Chen notation to create a diagram that was later implemented in a PostgreSQL instance. Moreover, I added data to be able to test advanced queries such as OLAP, Triggers and Window functions. Finally, I used Python and Tkinter to implement a Graphical User Interface (GUI) in which the user is able to perform C.R.U.D operations and advanced queries to explore the dataset.
In this class, my group worked on analyzing a heart disease data set to classify between patients with heart disease and patients without, we used a variety of method such as LDA, LASSO , KNN, Bagging, Random Forests, and Boosting. The best perfoming method Boosting reached an accuracy of 87.7%.
This project is a econometrics analysis of the video game market, it uses several variables such as critic rating, user rating, platform, and console on a multiple linear regression model in order to determine whether those variables influence video game sales.
Used Python (PyTorch/Pandas), Reinforcement Learning, and Deep learning algorithms to train a cryptocurrency trading agent that can buy, sell, and hold cryptocurrencies. The agent successfully achieved profits in the test dataset.
Papers that used and compared different time series forecasting methods such as Univariate models (ARIMA), Vector Autoregression models (VAR), and Vector Auto Correction models (VECM) to forecast cryptocurrencies prices using R and packages.
In this project we used PCA to reduce the dimension of the MRI images, after we achived the Image reduction we used classifications methods to classify between tumor and non-tumor images.
For further inquiries, or if my resume is needed please contact me using linkedin or email.