What I Build
Systems that predict behavior and drive decisions.
Churn Prediction
Predict which customers will leave before they do. LightGBM, XGBoost, and uplift modeling to target retention spend where it matters.
Revenue Optimization
Customer lifetime value modeling, ROI simulators, and budget optimization. Turn model outputs into dollar figures stakeholders understand.
Decision Systems
NLP pipelines for ticket routing, automated reporting infrastructure, and production ML systems. Built for speed and reliability.
Portfolio
A glimpse of the projects I've been working on
SignalForge
Churn prediction with statistical rigor. Optuna-tuned LR at 0.849 AUC with 5-fold CV and bootstrap CIs. Quantified $1.67M revenue at risk and built a retention ROI framework. The interesting finding: simple models beat ensembles.
Ticket Intel
NLP pipeline for routing support tickets, summarizing threads, and extracting entities. Built for speed — 12ms p99 latency, 500+ req/sec. Uses TF-IDF + Naive Bayes instead of LLMs because 'refund please' doesn't need GPT-4 to classify. 90% F1 across 5 categories.
SaaS Churn Simulator
Predicts which customers will churn and which retention offers work. LightGBM on 2.7M events achieves 0.85 AUC and 3.2x lift. Live ROI calculator shows budget impact before you spend.
AutoModeler
Auto-generates 3-statement financial models from public data. Pulls from Financial Modeling Prep API, projects 5 years forward, outputs linked Excel with native formulas. Cash flow reconciliation and balance sheet balancing built-in.
Experience
My professional and educational journey
Business Intelligence Analyst
Community Hospital
2022 - Present
Key Impact
automated clinical chart audits, reducing review time by 75%
direct to C-suite for operational decision-making
monitoring suicide risk, SDOH, wRVU, and quality metrics
designed and implemented clinical quality reporting workflow
Education
Dual MBA & M.S. Data Science
Eastern University
Expected 2027Bachelor of Applied Science
Peru State College
2022Recent Activity
Continuous learning and shipping.
Latest Commits
Loading commits...
Current Focus
ML Trading Models
Training predictive models for futures trading on CME
Quantitative Research
Backtesting systematic strategies with walk-forward validation
Production ML Pipelines
Real-time inference systems for market analysis
Open Source
Quant tools and data science utilities