Data Analyst & Engineer specializing in statistical modeling, data pipelines, and revenue operations.
Companies often have data trapped in different 'silos' (different software tools that don't talk to each other). I built Silobreak, a web application that pulls data from these different sources into one place, making it easy to analyze and visualize the big picture.
| Role | Full Stack Developer |
|---|---|
| Stack | React, Node, PostgreSQL |
| Status | Live |
While a full-time student, I taught myself how to code and built a complete online marketplace for the wine industry from scratch. It connects buyers and sellers, handling everything from user accounts to the actual transaction mechanics.
| Role | Founder & Lead Dev |
|---|---|
| Type | Two-Sided Marketplace |
| Status | Launched |
When a website tests a new design (A/B testing), they usually use confusing math (p-values) to guess if it worked. I used a more advanced method (Bayesian statistics) that gives a straight answer: 'There is a 97% chance the new design is better, and it will make us an extra $36,000 a year.'
| Method | Bayesian Inference |
|---|---|
| P(B > A) | 97.2% |
| Exp. Lift | +$36k/yr |
At Chalice AI, the sales team's call recordings (Gong) weren't matching up with their customer database (HubSpot). This meant missing data and broken reports. I built a system to automatically match and sync these records, ensuring management could actually trust their pipeline data.
| Role | BI & Sales Engineer |
|---|---|
| Impact | Resolved Attribution |
| Stack | Python, APIs, HubSpot |
Power companies need to know exactly how much electricity people will use next year so they don't run out. I analyzed 8 years of historical energy data and built a mathematical model that accurately predicts future energy demand, accounting for summer AC spikes and winter heating.
| Model | SARIMA |
|---|---|
| Test MAPE | 4.7% |
| Horizon | 12 Months |
Instead of just guessing *if* a customer will cancel their subscription, I used medical statistics (normally used to predict patient survival rates) to predict *when* they will cancel. I found that getting customers on annual contracts cuts the risk of them leaving by almost half.
| Model | Cox PH |
|---|---|
| Key Finding | Annual cuts churn 47% |
| N | 2,847 |
I built a machine learning model that acts like an automated real estate appraiser. By feeding it data on thousands of houses (square footage, quality, age), it learned how to accurately predict the sale price of a home, and can explain exactly which features added or subtracted value.
| Model | XGBoost |
|---|---|
| RMSE | $16,840 |
| R² | 0.947 |
I analyzed the chemical makeup of thousands of wines to figure out what makes a 'good' wine. By using advanced grouping techniques, I found that you can accurately predict a wine's quality score just by looking at its alcohol content and acidity levels, without ever tasting it.
| Method | PCA & Clustering |
|---|---|
| Model | Random Forest |
| Accuracy | 82% |