Healthcare AI · Fractional CTO · LLM & agent systems
Twenty years building HIPAA-grade software. Now I build LLM and agent systems that replace your rules engines, automate clinical operations, and survive a SOC 2 audit. Quietly. With evaluations. With governance.
20 years building healthcare software, end to end
The gap
The healthcare companies that need AI most are the ones that can least afford a wrong move. HIPAA. SOC 2. HITRUST. Auditors who read your prompts. Clinicians who lose trust on the second bad output. Boards that wanted a launch last quarter.
The cheap version of an AI consultant ships a Zapier graph and a ChatGPT screenshot. That isn't a system. It's a demo that will fail an audit and a clinical review on the same day.
I do the senior version of the work. Skill files. MCP servers. Agent orchestration. Evaluation harnesses. Governance. The same practice I just used to pivot a clinical-decision-support company from a rules engine to an LLM-driven pipeline — done for yours.
Services
Pick the one that fits the shape of your problem. Both publish their pricing because that's what senior work does.
Productized engagement
I design, build, and deploy a production-grade AI workflow inside your business. You keep the source.
Recurring engagement
An experienced healthcare CTO embedded in your leadership team for a fraction of full-time cost.
HIPAA, SOC 2, and HITRUST aren't add-ons. They're the starting position. Every system is designed to be audited.
Athena Marketplace approval. Clinical decision support in front of real clinicians. Workflows that have to clear a release gate.
Not a prompt-engineer-for-hire. A 20-year CTO who has built and led engineering at four healthcare companies.
Selected work
Head of Engineering · Jul 2025 — present
Replaced a clinical rules engine with an LLM and agent pipeline.
Chief Technology Officer · Aug 2023 — Jul 2025
Built the platform, data, and engineering org for a behavioral-health company.
Senior Director of Software Engineering · Nov 2022 — Aug 2023
Scaled a 30+ engineer org while pursuing HITRUST and Medicare interoperability.
Chief Technology Officer · Feb 2013 — Nov 2022
Built the IT organization from scratch across nearly a decade.
Insights
Compliance · 4 min
Most teams designing healthcare AI start with the model. The right starting point is the audit. Here's what actually shows up on a HIPAA control review for an LLM-powered workflow — and the design choices that make it pass cleanly.
Read
Architecture · 4 min
Rules engines are brittle, slow to extend, and expensive to test. LLMs are flexible, fast to extend, and impossible to test the same way. Here's the architecture pattern that lets you trade the first set of problems for the second — without losing your safety story.
Read
Integration · 4 min
Athenahealth's Marketplace approval process is a serious gate, especially when Claude or another LLM is in the data path. Here's the workflow that gets you through it without burning weeks on rejected submissions.
Read
Next step
Thirty minutes. No deck. No pitch. We'll look at one workflow together and decide whether it's a fit. If it isn't, you'll leave with a clearer picture of what is.