Master's student in Health Informatics at East Carolina University, focused on hands-on learning and preparing for an entry-level IT role.
A collection of class projects, labs, and practice work from my master's program that demonstrate the technologies and problem solving Im learning.
Analyzed a 500-person survey dataset to examine how social media usage predicts stress, sleep quality, and happiness, using Excel (validation, pivot tables), SQL/SQLite (aggregation), and Python (pandas, seaborn, scikit- learn). Built a Random Forest model predicting stress levels (R² = 0.517, MAE =
.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 in Health Informatics at East Carolina University, focused on hands-on learning and preparing for an entry-level IT role.
I am Krysten Jefferson, a graduate student in Information Technology at East Carolina University pursuing a master's degree. I have limited professional experience and am preparing to begin my career while building skills and academic projects in IT.
Externships
Pfizer Advanced: AI-Powered Document Insights & Data Extraction Externship
In progress
Experience
Health Insurance Agent — Balance Benefits Group
Balance Benefits Group · 2023 – 2025
Patient Advocate — ChartSpan
ChartSpan · 2023 – 2024
Remote Call Center Representative — NexRep (Teladoc Health)
NexRep (Teladoc Health) · 2020 – 2023
Guest Advocate, Women's & Kids' — Target
Target · 2018 – 2020
Education
East Carolina University
M.S. in Health Informatics and Information Management · Class of 2028
University of Maryland
B.S. in Psychology, Minor in Data Science · Class of 2023
Skills
Analyzed a 500-person survey dataset to examine how social media usage predicts stress, sleep quality, and happiness, using Excel (validation, pivot tables), SQL/SQLite (aggregation), and Python (pandas, seaborn, scikit- learn). Built a Random Forest model predicting stress levels (R² = 0.517, MAE =
Analyzed a 500-person survey dataset to examine how social media usage predicts stress, sleep quality, and happiness, using Excel (validation, pivot tables), SQL/SQLite (aggregation), and Python (pandas, seaborn, scikit- learn). Built a Random Forest model predicting stress levels (R² = 0.517, MAE = 0.87) and translated findings into summaries for non-technical audiences.
✅ 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.
This externship built a prototype pipeline for extracting structured insights from enterprise PDF documents using OCR, large language models, and retrieval-augmented generation. The work surveyed LLM fundamentals, designed extraction and search flows, and produced demonstrable artifacts for processing documents and surfacing passage-level answers.
I examined a vendor pharmaceutical document and documented concrete AI-readability issues, including misread date formats, form-version inconsistencies, and example parsing errors that could affect downstream extraction.
A sample vendor document was examined to assess AI-readability. I compared date formats and form versions, noted OCR could misread dates and struggle with legacy layouts, and recorded specific examples of potential parsing errors.
