AI Engineer· Data Scientist · Epidemiologist
MPH-trained, production-hardened data scientist building at the intersection of public health domain expertise and LLM engineering.
My background is unusual in AI engineering circles. I trained as an epidemiologist, spent years doing applied NLP and ML in a real production public health context, and I'm now building toward LLM-powered systems and agent architectures. That domain depth — knowing what syndromic surveillance actually means, understanding why a false positive in a suicide ideation classifier has real downstream consequences — shapes how I think about model design and data systems in ways that purely technical backgrounds often don't.
At the Connecticut Department of Public Health, I've worked across Syndromic Surveillance and COVID-19 response. My early work involved building and refining NLP-driven syndrome classifiers on near-realtime emergency department data: opioid overdose patterns, suicidal ideation (with intentionality modifiers), suspected carbon monoxide poisoning, and more. During the pandemic, I moved into daily reporting pipelines feeding the Governor's office and the CT Open Data Portal — and helped build out the automation and data engineering infrastructure that DPH still runs on today.
Now I'm pushing deeper into the engineering side of AI: LLM APIs, agent orchestration, and production-ready tooling. I'm a committed FOSS advocate — the same principles of openness and reproducibility that make great science make great software, and I try to bring both to everything I build.
- 🤖 Working on my homelab and building out my local AI stack
- 📖 Day-to-day stack: R · Python · SQL
Languages
AI / Data
DevOps / Infra
