Systems I'm shipping at work, capstones from SavvyCoders and TripleTen, and the full 16-project data science portfolio. Each card is honest about what it tried to do, what came out, and the stack that got it there.
The internal operating system behind a 170-project, 15-state commercial solar portfolio. Each one replaces a spreadsheet, a folder of PDFs, or a manual QuickBooks workflow.
Solar-industry operations and business intelligence platform in active development. Combines a vendor/product reference database, regulatory eligibility tracking, and ERP-style modules covering customers, jobs, products, schedules, and logistics — all built around a structured ID + schema system.
Hub-and-spoke architecture using Claude as the primary orchestrator across specialized agents for code generation, persistent memory, and desktop automation. Configured through a versioned CLAUDE.md with explicit routing rules, role declarations, and prompt contracts between agents.
Progressive fuzzy-match engine across 5 bank accounts in a high-volume, multi-account environment. Tier-priority confidence scoring with review queues for ambiguous matches — surfaces discrepancies before month-end, not after.
Programmatic IIF generation for customers, POs, invoices, and deposits with anti-duplicate safety checks against tens of thousands of historical transactions.
pandas + openpyxl pipeline covering P&L, cash flow, billing, margin, and schedule. Auto-regenerates; static HTML operations hubs with collapsible sections and clickable file links.
OCR + classification + auto-routing for hundreds of incoming vendor docs, bank statements, and internal cost models. Outlook .msg / .eml forensics pipeline builds defensive timelines for dispute preparation.
Working reference system for tracking federal incentive eligibility, manufacturer sourcing rules, and domestic content thresholds across product categories. Built to help a commercial solar team make faster, more defensible procurement decisions.
SavvyCoders Capstone
Full-stack Movie Data Single Page Application. Users explore trending films, upcoming releases, and box-office insights via real-time data from the TMDB API. Designed the UI, built client-side routing, reusable components, and async API integration end to end.
Personal Project
A dairy-free ingredient checker built as a Progressive Web App. Lets you scan or type ingredients to spot dairy hiding in product labels — with a flashlight for reading small text, confetti for clean hits, and PWA install.
TripleTen Program
Computer-vision regression model for alcohol-sales compliance — estimates age from a photograph using a CNN trained on a labeled image dataset. Built for the supermarket chain Good Seed to help prevent sales to underage customers.
Personal Site
This site — an Astro static build deployed on Netlify. A place to write, log what I'm learning, and ship small web work.
The end-to-end DS portfolio: churn, pricing, forecasting, NLP, computer vision, and classical analytics. Full code and notebooks in the TripleTenProgram repo.
Predicted contract cancellations so marketing could reach at-risk customers before they churned.
AUC-ROC ≥ 0.75
Random Forest · Logistic Regression · CNN · TensorFlow
Gradient-boosted model estimating used-car market value with a focus on prediction speed as well as accuracy.
Fast tuned GBM beat baseline linear
CatBoost · LightGBM · XGBoost · scikit-learn
NLP pipeline flagging negative movie reviews for sentiment tracking.
Multi-model NLP with transformer features
NLTK · spaCy · transformers · LightGBM
Hourly forecast of taxi demand to help optimize fleet scheduling.
RMSE ≤ 48 on hourly forecast
statsmodels · scikit-learn · LightGBM
Classification model that picks the best cell plan for a customer from usage patterns.
Accuracy ≥ 0.75
scikit-learn · Decision Tree · Random Forest · Logistic Regression
Predicted which bank customers were likely to leave so retention could be targeted.
F1 ≥ 0.59
scikit-learn · Random Forest · Logistic Regression
Evaluated borrower metrics to predict the likelihood of loan default.
Identified drivers of default risk
Python · pandas
Validated reserve volume models and calculated profit/risk trade-offs across candidate drilling regions.
Best risk-adjusted region selected
scikit-learn · Linear Regression · Bootstrap
Modeled the gold-mining recovery process to surface efficiency improvements.
Prototype model for industrial use
scikit-learn · Linear Regression · Decision Tree · Random Forest
Identified similar customers and predicted insurance benefit amounts while preserving data privacy.
Privacy-aware similarity + benefit model
scikit-learn · KNN · seaborn
Investigated which features drive used-vehicle prices in classified ads.
Feature-importance report for pricing
pandas · NumPy · Matplotlib
Tested hypotheses on user vs. critic scores to pick promising platforms and ad directions.
Statistically validated recommendations
pandas · SciPy · Matplotlib
Tested whether weather conditions meaningfully shift taxi trip duration.
Hypothesis test on real trip data
pandas · SciPy
Analyzed client behavior across telecom packages to identify which plans generate the most revenue.
Revenue-driver segmentation
pandas · SciPy · Matplotlib
Compared listening habits between two cities to surface patterns in music preference.
First end-to-end EDA project
Python · pandas
End-to-end DS workflow — cleaning, modeling, evaluation — focused on performance metrics and outcome prediction.
Full workflow, performance metrics reported
Python · scikit-learn · pandas