As a research software developer, I tackle exciting challenges in computational biology and bioinformatics. I primarily work analyzing spatial transcriptomics, single-cell, and imaging data. My interests lie in developing interpretable, open-source machine learning, statistical, and computational software that empowers researchers to uncover meaningful insights from large-scale, high-dimensinsional biological & biomedical data, ultimately contributing to a deeper understanding of disease, development, and human health. I'm committed to creating user-friendly and accessible tools with strong data visualization principles and plan to pursue a Ph.D. to advance novel methodologies for the broader scientific community.
A showcase of my work in machine learning, deep learning, computer vision, computational biology, full-stack development, and software development
An R and Python pipeline that converts imaging-based spatially resolved transcriptomics (im-SRT) data into image-based tensor representations to be used with feature extractors (ResNet101) for further downstream analysis in the identifcation and characteriztion of novel cell types & states by capturing subtle subcellular RNA heterogeneity that traditional gene count methods often miss.
An R package designed for visualizing proportional data across spatial coordinates through scattered stacked bar plots. Built with ggplot2, dplyr, and tidyr, offering customizable visualizations for spatial data analysis and published on GitHub/CRAN.
A machine learning model built using PyTorch to predict COVID-19 hospitalization and ICU admissions based on CDC surveillance data. Achieved a 94.4% PR AUC score with a Random Forest classifier through hyperparameter optimization and 3-fold cross-validation.
A real-time drowsiness detection system using OpenCV, Mediapipe, and dlib for facial landmark recognition. Analyzes Eye Aspect Ratio (EAR) and Mouth Aspect Ratio (MAR) to classify driver states, achieving 70.23% accuracy.
A full-stack chatbot developed to provide accessible cancer education to patients and families. Built with HTML, CSS, JavaScript, and React, featuring dynamic chatboxes, response delay handling, and accessibility improvements.
A collaboration with Universidad de Madrid exploring the Lottery Ticket Hypothesis through genetic algorithm development for optimizing supermasks in convolutional neural networks, achieving 8.5% performance improvements over random guessing.
Research contributions in computational biology and bioinformatics
Programming languages, packages, toolkits, software, and systems I work with and am proficient in
Download my full resume for detailed information about my experience and qualifications
Software Developer & Post-Baccalaureate Researcher
PDF • Updated December 2025B.S. Computer Science & Chemical Engineering
Johns Hopkins University
Software Developer
JEFWorks Lab
Computational Biology
Spatial Transcriptomics
Computer Vision
Deep Learning
I'm always interested in discussing new opportunities and collaborations! (◕ っ ◕✿)
