I BuildProduction ML

I am an ML Engineer specializing in production machine learning systems for high-stakes healthcare environments. My work sits at the intersection of predictive modeling, data engineering, and clinical impact - building systems that don't just process data, but support decisions that affect real lives.

My core focus is end-to-end ML pipeline architecture: from raw data ingestion through Medallion Architecture (Bronze → Silver → Gold) to recall-optimized models in production. I build with XGBoost, Python, and SQL Server, and I care deeply about the unglamorous fundamentals - data quality, patient-isolated splits, drift detection, and threshold engineering that holds up in the real world.

I approach healthcare ML with a bias toward reliability over novelty. Whether designing fall risk prediction systems, multi-criteria scoring frameworks, or audit pipelines for clinical datasets, I optimize for what matters in regulated, human-facing environments: interpretability, reproducibility, and measurable outcomes.

Before specializing in ML, I built across the full stack - agentic AI workflows with LangChain and RAG, pixel-perfect frontends in Next.js, and production React Native apps on AWS. That breadth shapes how I think: I design ML systems the way a full-stack engineer would - with the entire data-to-decision chain in view.

I'm driven by problems where engineering precision has direct human consequence. My goal is to build adaptive, auditable systems that reason over uncertainty, degrade gracefully, and deliver impact you can measure - not just demonstrate.