Data Analyst · Toledo, Ohio

Data that
serves people.

I turn messy operational data into clear insight for mission-driven teams — with a background in community health, UX design, and field operations shaping every analysis.

Current Stack

Python SQL pandas scikit-learn NumPy Streamlit matplotlib Git Jupyter

Focus Areas

Healthcare Analytics Program Evaluation Predictive Modeling Data Storytelling
700+
Patients supported at Mercy Health
5+
End-to-end data projects
About

A little background

I'm a data analyst building toward work that actually matters. My path here wasn't a straight line — I came through community health navigation, conversational UX design, and field operations — and that background shapes how I approach data: practically, with real-world context, and always with people in mind.

I use Python and SQL to clean datasets, explore patterns, build predictive models, and create visualizations that help teams make confident decisions. I'm currently completing a Data Science & Machine Learning certificate at TripleTen (expected April 2026), where I'm deepening my applied analytics and modeling work.

I'm most interested in roles where data supports research, program evaluation, or mission-driven innovation — especially in healthcare, education, and community-focused organizations. I care less about impressive-looking dashboards and more about insights that actually inform decisions for people who need them.

Projects

Selected work

01

Customer Retention — Churn Prediction

Organizations lose members, patients, and students before they ever ask for help. Built a classification model to identify who's at risk early — so teams can act before disengagement becomes departure. Addressed real-world class imbalance, optimized for precision-recall balance, and focused the output on decision support rather than model performance for its own sake.

Python scikit-learn Classification Imbalanced Data Feature Engineering
View on GitHub →
02

Taxi Service — Demand Forecasting Model

When services are under-resourced, the people who depend on them most feel it first. Built a time-series forecasting model using historical demand patterns to anticipate peak usage periods — giving operations teams the information they need to plan staffing, allocate resources, and improve access before problems emerge. Hit an RMSE target of ≤ 48.

Python pandas Time Series Regression Feature Engineering
View on GitHub →
03

Community Feedback Analysis — NLP Text Classification

Most organizations collect more qualitative feedback than they can ever manually read. Built an NLP pipeline to classify and surface patterns from large volumes of unstructured text — turning open-ended responses into structured insight. Achieved F1 above 0.85 across multiple model approaches. Applicable anywhere teams need to understand what people are actually saying: surveys, program evaluations, community input.

Python NLP Text Classification scikit-learn Gradient Boosting
View on GitHub →
Skills

Tools & methods

Languages

  • Python
  • SQL
  • HTML / CSS

Libraries & Tools

  • pandas · NumPy
  • scikit-learn
  • matplotlib · Plotly
  • Streamlit
  • Jupyter · Git

Analysis & Modeling

  • Exploratory Data Analysis
  • Classification & Regression
  • Time Series Forecasting
  • Statistical Hypothesis Testing
  • Feature Engineering

Domain Knowledge

  • Healthcare Operations
  • Program Evaluation
  • Workflow Analysis
  • Human-Centered UX
  • Community Organizing
Contact

Let's talk

I'm currently open to data analyst roles at mission-driven organizations — particularly in healthcare, education, or community-focused work. If you're building something that matters and need someone who can make data useful, I'd love to hear from you.

The goal is to turn data into information, and information into insight.

— Carly Fiorina