I turn messy business problems into AI systems that run in production. Clinical lab scientist turned engineer.
About
I start by understanding how the work actually gets done. Then I integrate with the systems a team already uses, deploy the solution, and measure whether it helps.
I came into this from healthcare, with a BSc in Medical Laboratory Science. That background shapes how I build: software has to fit the real workflow and hold up when people depend on it.
Some examples: voice AI agents that triage and dispatch calls for two plumbing companies, ClaimGuard, which automates insurance denial appeals with a LangGraph pipeline, and SESKit, an open source developer email platform on AWS SES. LabLens, a hackathon build, is an MCP server that gives Claude access to Epic's FHIR sandbox to summarize and flag lab results.
Projects
Six projects that show how I work. Smaller builds and demos are on the projects page.

AI Content Engine with Approval Flow (fencing contractor)
A fencing contractor needed regular SEO content without the owner logging into a CMS.
Zapier agents draft a post every weekday, and the owner approves or rejects it by email before anything publishes.
Work Experience
Skills
AI and agents: Claude API, OpenAI API, LangGraph, MCP servers, Voice AI (Vapi)
Backend: Python, FastAPI, TypeScript, PostgreSQL, Redis
Infra: AWS, Docker, Terraform, GitHub Actions, Railway
Integrations: n8n, Zapier, Webhooks, Epic FHIR
Contact
Want to chat? Send me an email or connect with me on LinkedIn and I'll respond whenever I can.
Send me an email


