Data Science · Machine Learning · Analytics
Turning messy data into decisions that actually move the needle.
M.S. Data Science, Pace University (GPA 3.86). 4+ years building ML systems and BI infrastructure across healthcare and manufacturing. Authorized to work in the U.S. without sponsorship.
Pace University
audit · zero findings
experience
Selected work
Things I've built
Real data. Real models. Explainable outputs.
↗ hover cards to feel the tilt
Featured · Mar 2026 – Present
AI-Powered Decision Intelligence Platform
AWS · LangChain · FAISS · MLflow · dbt · Airflow · Snowflake
End-to-end analytics on AWS — raw data → S3 → dbt → Airflow across millions of rows.
Classification and regression in Python; MLflow for experiments; SHAP for non-technical stakeholder explainability.
LangChain + FAISS vector search so users can ask natural-language questions over their business data.
Deployed on EC2 with Docker + FastAPI backend, Streamlit front-end, Snowflake storage.
Oct – Dec 2025
FairLens AI
Product Sustainability Scorer
XGBoost + GPT-4 RAG pipeline generating health-risk scores from barcode data. 30+ features, ~85% accuracy. Fairness audits with AIF360.
2025
AI Task Recommendation Agent
Gemini · Kaggle Agents
Digital twin agent that adapts scheduling to user context and energy levels. Multi-agent reasoning with rule-based prioritization.
2025
MoMA Exhibition Helper
Scikit-learn · SHAP
Logistic regression predicting artwork exhibition selection. Engineered features: medium, period, artist history. SHAP for probability interpretation.
2024
Audio Emotion Recognition
SVM · Librosa · MFCC
SVM classifier on MFCC features from speech clips — mapping voice signals to emotion states. Confusion matrix analysis to identify clashing pairs.
Experience
Where I've worked
Real impact across healthcare operations, manufacturing analytics, and independent consulting.
- Built and deployed classification and regression models in Python for small business clients — presented through dashboards and plain-language summaries for non-technical decision makers
- Completed Azure-based ML project deployments end-to-end, from model training through production workflow setup
- Guided 5+ college students through complete data science projects covering problem framing, feature engineering, model selection, and interpretation
- Digitized fully paper-based records — built attendance tracking, billing validation, and P&L reporting systems that cut weekly manual reporting from ~6 hours to under 2
- Identified seasonal attendance patterns and translated findings into staffing and budget forecasts leadership could act on directly
- All 4 facilities passed state inspection with zero findings
- Built Looker dashboards for ops managers, floor supervisors, and senior leadership — replacing a fragmented manual Excel process requiring constant cross-team coordination
- Python + SQL EDA to surface production bottlenecks: idle machines, stage dependencies, inventory mismatches
- Time-series forecasting (rolling averages, seasonality decomposition) to align production schedules with demand and reduce overstock
Skills
What I bring
Where human intuition
meets machine intelligence.
Languages & data
Databases
ML & AI
Cloud & tools
Visualization
Communication & people
Credentials
Education & certifications
Get in touch
Let's work on something real.
Open to full-time roles in data science, ML, and analytics. If there's a project where focused ML work could make a clear difference — I'd love to hear about it. I usually reply within 1–2 business days.