Malcolm Zartman

Health-tech startup using multi-dimensional data to deliver personalized insights for longevity and well-being.

1. Data Scientist

   May 2026 – Present
   • Built predictive XGBoost models on an automated feature selection pipeline (RFECV, LightGBM, SHAP), reducing 20,000+ clinical variables by over 95% to key predictors across 500,000+ records, accelerating domain research and improving health-scoring precision
   • Engineered a modular, HIPAA/NIH-compliant ETL pipeline in Python (Pandas, NumPy, SciPy) with schema normalization, provenance tracking, feature ablation testing, and automated integration tests, cutting dataset integration time from 2 weeks to minutes
   Show moreShow less

   • Replaced static threshold cutoffs with continuous actuarial hazard models, enabling more granular and defensible risk estimates
   • Developed interactive dashboards (Streamlit, Plotly, Excel) and documentation that translate complex datasets, internal research, and model architecture into clear insights for non-technical leadership
   • Built an insights engine that surfaced a novel clinical finding consistent with published research, informing product strategy and investor decks
   • Earned an executive commendation for onboarding cross-functional technical talent and clinical experts, establishing HIPAA/NIH data governance standards
   • XGBoost
   • RFECV
   • LightGBM
   • SHAP
   • Pandas
   • NumPy
   • SciPy
   • Streamlit
   • Plotly

2. Data Scientist Intern

   Dec 2025 – May 2026
   • Architected a proprietary, pathology-agnostic health-scoring engine and ranking algorithm from scratch on 200,000+ clinical records, synthesizing multi-modal data into actionable longevity insights, establishing the company's data science foundation
   • Delivered a full-stack MVP in under 30 days for investor and partner demos: a containerized Flask REST API (Docker) with a React.js/Tailwind CSS frontend, configured CI/CD via GitHub Actions, and deployed on Azure and Render
   Show moreShow less

   • Wrote Databricks SQL to aggregate multi-source clinical data and profiled patient risk using K-Means, optimizing cohort granularity with PCA and silhouette analysis
   • Applied temporal sequence mining to surface drug-interaction patterns consistent with published research
   • Partnered with the CTO and Product Manager on technical roadmaps, presenting statistical modeling and analytics to leadership to support seed-investor engagement
   • Flask
   • Docker
   • GitHub Actions
   • Azure
   • Databricks SQL
   • K-Means
   • PCA
   • React.js

AgeTech non-profit supporting seniors and their caregivers through digital tools.

1. Software Developer Intern

   May – Jul 2024
   • Engineered a multi-source data ingestion pipeline in Python (BeautifulSoup) that automated catalog scraping across 10+ partner networks into structured Excel reports, establishing the organization's first lead database and enabling strategic collaborations and a new revenue stream.
   • Architected a responsive WordPress web platform using HTML, CSS, and JavaScript in Agile sprints, prioritizing accessibility and navigation for seniors and caregivers while serving as the technical foundation for the organization's mobile application launch
   Show moreShow less

   • Designed and implemented the organization's first affiliate marketing framework within the application, providing the technical infrastructure to scale partnerships and support long-term sustainability
   • Translated high-level business goals into technical features during Agile development, ensuring the platform remained highly accessible for both seniors and caregivers while improving navigation and discovery time from days to minutes
   • Python
   • BeautifulSoup
   • Excel
   • WordPress
   • HTML
   • CSS
   • JavaScript
   • Agile