Master's student at Nova Southeastern University focused on healthcare data analysis, public datasets, and AI-assisted tools to strengthen my resume.
Portfolio of practical work that uses public health datasets and AI-assisted tools to analyze infections, outcomes, and access measures, completed during a TruBridge externship and academic study.
.jpeg)
TruBridge · Jun 2026 · ✅ Verified by Extern
A TruBridge externship on healthcare data analytics, focused on using public datasets and AI-assisted tools to analyze infections, outcomes, and access measures.
Conducted exploratory data analysis (EDA) on county level CDC PLACES public health data using Python (Google Colab) to identify relationships between diabetes prevalence, healthcare access, and geographic disparities. Created Tableau visualizations and a Gamma presentation to communicate county level healthcare insights to non-technical audiences. Developed an interactive Claude Artifacts dashboard by integrating research findings, healthcare datasets, and interactive visualizations. Leveraged ChatGPT, Claude, and prompt engineering to streamline research, technical writing, dashboard development, and analytical workflows.
Developed a Bidirectional LSTM model using TensorFlow/Keras for sentiment analysis on the IMDB dataset, achieving over 85%+ validation accuracy. Evaluated model performance using precision, recall, F1-score and ROC-AUC. Fine-tuned a pre-trained GPT-2 model using Hugging Face Transformers to generate synthetic text data, applying tokenization and sequence padding for preprocessing. Generated and analyzed 100+ synthetic data points to asses sentiment predictions and model output quality. Visualized model performance and probability distributions using Python (Matplotlib).
Developed and trained a CNN using Python and TensorFlow/Keras to classify images across 10 categories in the CIFAR-10 dataset. Achieved over 87%+ validation accuracy through model optimization techniques including Batch Normalization and Dropout. Evaluated model performance using classification metrics such as precision, recall, and F1-score, and Analyzed misclassified images to interpret prediction errors. Visualized training and validation using Matplotlib.
Designed and implemented a relational database system using MySQL to manage streaming platform data (movies, users, subscriptions, watchlists, etc). Developed a normalized schema from an Entity Relationship (ER) model supporting one-to-many and many-to-many relationships. Wrote and executed complex SQL queries (multi-table joins, aggregate functions, subqueries, GROUP BY, HAVING) to analyze user and content data. Built and validated the database with structured sample data, ensuring data integrity through constraints and key relationships.
Developed a CGAN using Python, TensorFlow, and Keras to generate synthetic images of clothing items from the Fashion MNIST dataset. Designed and implemented both generator and discriminator models to learn and replicate image distributions across labeled classes. Preprocessed image data using normalization and one-hot encoding to prepare inputs for model training. Evaluated model performance and visualized generated outputs using Matplotlib to assess image quality and training progress.
.jpg)
Pfizer · ✅ Verified by Extern · ⏱️ In progress
Prototype AI-powered document intelligence with Pfizer—using OCR, LLMs, and RAG to automate real enterprise PDF workflows and build a standout portfolio project.
Master's student at Nova Southeastern University focused on healthcare data analysis, public datasets, and AI-assisted tools to strengthen my resume.
I am a Master of Science in Information Systems student at Nova Southeastern University with a BS in Civil Engineering. I have experience as an AI trainer/reviewer and as a research assistant. My work focuses on data analysis, visualization, databases, and machine learning using Python, TensorFlow/Keras, SQL, Tableau, and Power BI. I have completed healthcare data analytics and multiple deep learning and database projects, and I am strengthening my resume to move into software and AI roles.
Externships
Pfizer Advanced: AI-Powered Document Insights & Data Extraction Externship
Pfizer
Experience
AI Trainer/Reviewer -Software Engineering/Coding Domain
Handshake AI Fellowship · 06/2026 – Present
Research Assistant
Transportation Engineering Research Lab · 01/2022 – 05/2023
Education
Nova Southeastern University
Master of Science in Information Systems
Western Kentucky University
Bachelor of Science in Civil Engineering · Class of 2023
Skills
✅ Verified by Extern
A TruBridge externship on healthcare data analytics, focused on using public datasets and AI-assisted tools to analyze infections, outcomes, and access measures.
The work investigated county-level diabetes and healthcare access using public sources, assembling and cleaning CDC PLACES records and related datasets, conducting exploratory and correlation analyses, and building interactive visualizations. The work was awarded Top Performer by TruBridge.
I produced cleaned, measure-specific subsets from CDC PLACES, performed EDA and correlation analysis on Tennessee county data, and delivered an AI-assisted interactive dashboard plus a recorded presentation summarizing geographic disparities and findings.
Conducted exploratory data analysis (EDA) on county level CDC PLACES public health data using Python (Google Colab) to identify relationships between diabetes prevalence, healthcare access, and geographic disparities. Created Tableau visualizations and a Gamma presentation to communicate county level healthcare insights to non-technical audiences. Developed an interactive Claude Artifacts dashboard by integrating research findings, healthcare datasets, and interactive visualizations. Leveraged ChatGPT, Claude, and prompt engineering to streamline research, technical writing, dashboard development, and analytical workflows.
Developed a Bidirectional LSTM model using TensorFlow/Keras for sentiment analysis on the IMDB dataset, achieving over 85%+ validation accuracy. Evaluated model performance using precision, recall, F1-score and ROC-AUC. Fine-tuned a pre-trained GPT-2 model using Hugging Face Transformers to generate synthetic text data, applying tokenization and sequence padding for preprocessing. Generated and analyzed 100+ synthetic data points to asses sentiment predictions and model output quality. Visualized model performance and probability distributions using Python (Matplotlib).
Developed and trained a CNN using Python and TensorFlow/Keras to classify images across 10 categories in the CIFAR-10 dataset. Achieved over 87%+ validation accuracy through model optimization techniques including Batch Normalization and Dropout. Evaluated model performance using classification metrics such as precision, recall, and F1-score, and Analyzed misclassified images to interpret prediction errors. Visualized training and validation using Matplotlib.
Designed and implemented a relational database system using MySQL to manage streaming platform data (movies, users, subscriptions, watchlists, etc). Developed a normalized schema from an Entity Relationship (ER) model supporting one-to-many and many-to-many relationships. Wrote and executed complex SQL queries (multi-table joins, aggregate functions, subqueries, GROUP BY, HAVING) to analyze user and content data. Built and validated the database with structured sample data, ensuring data integrity through constraints and key relationships.
Developed a CGAN using Python, TensorFlow, and Keras to generate synthetic images of clothing items from the Fashion MNIST dataset. Designed and implemented both generator and discriminator models to learn and replicate image distributions across labeled classes. Preprocessed image data using normalization and one-hot encoding to prepare inputs for model training. Evaluated model performance and visualized generated outputs using Matplotlib to assess image quality and training progress.
✅ Verified by Extern · ⏱️ In progress
Prototype AI-powered document intelligence with Pfizer—using OCR, LLMs, and RAG to automate real enterprise PDF workflows and build a standout portfolio project.
The externship prototyped AI-powered document intelligence for enterprise PDF workflows. The work combined OCR, retrieval-augmented generation, and large language models to process and analyze documents. Deliverables included research and technical artifacts that explained LLM mechanisms and their role in document extraction.
I produced a technical explanation of how large language models are trained, how they tokenize input, and how Transformer attention enables token-by-token prediction applied to pharmaceutical documents.
I explained how large language models were trained on vast text, how they tokenized input, and how Transformer attention guided token-by-token predictions to generate responses.
