Data Analyst & Founder
Python • SQL • R • Machine Learning • Tableau
I use data to find the mechanism behind the pattern — not just what happened, but why, and what it means for the decision in front of you. My work spans financial modeling, health analytics, and corporate strategy. The throughline is the same across all of it: I want to understand the system, not just describe it.
Currently: MSBA candidate at Georgetown University (December 2026) and founder of a health tech company in the Black maternal health space. The research that informs that work is public. The product is coming.
The Question: Black women in the U.S. die from pregnancy-related causes at 2–3× the rate of white women. The disparity holds across income, education, and geography. So what actually drives it — and what changes outcomes?
Approach: Full data science pipeline calibrated against CDC WONDER, KFF, and NVSS published figures. SQL relational schema → Python EDA → Random Forest, Gradient Boosting, and Logistic Regression → intervention simulation modeling.
Key Findings:
concerns_dismissedranks among the top predictors of adverse birth outcomes — above insurance type, above hospital level- Doula access for Black women on Medicaid: ~4%. For white women with private insurance: ~22%
- A what-if simulation combining advocacy improvements predicts a ~25% reduction in adverse outcome risk for Black women
- The model confirms: this is a systemic problem, not a clinical one
Why it exists: This project is the research foundation for something I'm building. More on that soon.
Skills: Random Forest • Gradient Boosting • Logistic Regression • SQLite • EDA • Health Analytics • Disparity Modeling • Intervention Simulation
Business Problem: Investment teams need to compare portfolio strategies on a risk-adjusted basis — not just raw returns, but the full picture including drawdown, volatility, and efficiency.
Solution: End-to-end financial analysis pipeline: SQL database layer for structured data management, Markowitz portfolio optimization to solve for min-variance and max-Sharpe portfolios, and a backtest engine comparing 5 investment strategies.
Key Findings:
- Max Sharpe portfolio achieved a 3.05 Sharpe ratio vs. SPY's 2.68 — optimized weights translated directly into real backtest outperformance (58.1% vs. 38%)
- Momentum strategy posted the 2nd-best return but the worst drawdown — clear illustration of return vs. risk tradeoff
- Min Variance vs. Max Sharpe allocations told opposite stories: one loaded PG for safety, the other tilted V and JPM for alpha. The efficient frontier made the tradeoff visual.
Skills: Markowitz Optimization • Portfolio Theory • Risk Metrics (Sharpe, Sortino, VaR, Max Drawdown) • SQL • Financial Modeling • Strategy Backtesting
Business Problem: Digital advertising budgets are wasted on audiences unlikely to engage on LinkedIn.
Solution: Predictive model using demographic data to identify high-probability LinkedIn users, enabling precise audience targeting.
Business Impact:
- 30–40% reduction in wasted ad spend through demographic pre-filtering
- Improved conversion rates via data-driven audience segmentation
- ROI optimization for B2B marketing campaigns
Skills: Logistic Regression • Feature Engineering • Interactive Web Apps • Marketing Analytics
→ View Project | → Try Live App
Business Problem: Asset managers and corporate boards need quantifiable ESG metrics to screen investments and benchmark climate commitments against peers.
Analysis: Multi-country regression analysis quantifying the relationship between emissions output and climate finance contributions. Built a scoring framework that flags misalignment between a company's emissions profile and its funding behavior.
Key Finding: Emissions explain only 40% of climate finance (R²=0.403) — the remaining 60% is driven by political and economic factors, which means standard emissions-based screening misses major risk signals.
Business Applications:
- ESG Investment Screening — Flag portfolio companies with weak climate commitments before they become liabilities
- Benchmarking — Compare corporate ESG performance against sector peers using a standardized scoring model
- Risk Assessment — Quantify climate transition risk across a holdings portfolio
- Regulatory Compliance — Track alignment with TCFD and CDP reporting standards
Skills: Regression Analysis • ESG Metrics • Risk Scoring • Data Visualization • Benchmarking
Business Problem: Organizations with finite budgets need a data-driven framework to identify where spending is misaligned with impact — before the next budget cycle.
Analysis: Cross-sectional analysis comparing resource allocation against quantifiable impact metrics across multiple categories. Built an efficiency scoring model that ranks allocation decisions by ROI.
Key Finding: Systematic funding gaps emerge when attention concentrates — some high-impact areas receive 80% less funding than their impact severity warrants. The model flags these before decisions are locked in.
Business Applications:
- Budget Optimization — Quantify ROI across program areas and reallocate toward highest-impact categories
- Grant & Investment Strategy — Identify underserved segments with the highest marginal return
- CSR Program Design — Build defensible allocation frameworks for corporate social programs
- Market Gap Analysis — Find underserved segments where investment has the most room to grow
Skills: Portfolio Optimization • Gap Analysis • Statistical Modeling • ROI Frameworks • Executive Dashboards
Business Problem: Corporations need to forecast whether current trajectories will hit committed targets — and if not, quantify exactly how much acceleration is required.
Analysis: Time-series trend analysis comparing actual performance trajectories against committed targets. Built a scenario model that outputs the required rate of change to hit each target by the deadline.
Key Finding: 4 of 5 entities tracked are off-target — some need to accelerate 2.7× to meet commitments. The model flags this early enough to course-correct.
Business Applications:
- KPI Forecasting — Project whether current trajectory meets stated targets before it's too late
- Scenario Planning — Model what acceleration looks like in practice: cost, timeline, resource requirements
- Supply Chain Risk — Identify high-risk partners based on their own trajectory data
- Investor Relations — Communicate credible performance roadmaps backed by data
Skills: Time Series Analysis • Forecasting • Scenario Modeling • KPI Tracking • Trend Analysis
| Status | Project | Focus |
|---|---|---|
| In Progress | Education Investment Efficiency Analysis | Spending benchmarking, outcome modeling, regression, clustering |
| Coming — April 2026 | [fenna] | A health tech product in the Black maternal health space. The research is here. The product is coming. |
Programming & Analysis:
- Languages: Python • R • SQL
- Analytics: Regression • Classification • Time Series • Clustering • Predictive Modeling • Portfolio Optimization
- Visualization: ggplot2 • Matplotlib • Seaborn • Plotly • Streamlit • Tableau • Interactive Dashboards
- Tools: Jupyter Notebook • SQLite • Git • GitHub
Business Domains:
- Financial Analysis — Portfolio optimization, risk-adjusted returns, strategy backtesting, financial modeling
- Health Analytics — Disparity modeling, outcome prediction, intervention simulation, equity research
- Marketing Analytics — Customer segmentation, audience targeting, conversion optimization, campaign ROI
- ESG & Investment Screening — Climate risk quantification, ESG scoring, regulatory benchmarking
- Corporate Strategy — Budget optimization, KPI forecasting, scenario planning, gap analysis
Key Data Sources:
- Finance & Business: Yahoo Finance, SEC filings, Kaggle, Bloomberg
- Health & Equity: CDC WONDER, KFF, NVSS, MMRIA, ACOG
- ESG & Sustainability: Climate Watch, OECD, Our World in Data
- Research & Benchmarking: Pew Research, Census Bureau, World Bank, UNESCO
I build analyses that translate directly into decisions — not just charts, but frameworks someone can act on.
What makes my work different:
- End-to-end ownership — From SQL database design to executive dashboards, I handle the full pipeline
- Risk-aware thinking — Whether it's portfolio drawdown, budget misallocation, or health outcome disparities, I model the downside, not just the upside
- Business-first framing — Every technical choice is driven by what the decision-maker actually needs
- Reproducible methods — Clean code, clear documentation, results that hold up when someone else runs them
Master of Science in Business Analytics | Georgetown University
McDonough School of Business | Expected December 2026
Merit Scholarship Recipient
"Data tells stories that matter — especially the ones hiding in plain sight."