Putting AI into real healthcare workflows
What actually got built in practice — what was done, how it landed, what results it produced, and where we took the wrong turn. Each case aims to show checkable evidence: real product interfaces, stage metrics, and the mistakes made along the way. Everything is redacted; the focus is method, judgment and boundaries.
Comorbidity Management: Routing Risk Alerts into Outpatient Referral and Inpatient Consultation
Built on a hospital's existing clinical systems, the platform uses confirmed clinical rules to flag comorbidity risks, delivers alerts to the physician workstation, outpatient text messages and specialist consultation, and records how physicians respond.
Consumer Healthcare Operations and Agent Management System
One FDE, a month and a half: products, billing, referrals, follow-up, and service delivery connected across five interfaces at a rehabilitation hospital.
RadIQ: From Radiology Reports to Quality Management and Clinical Collaboration
Voice-generated reports, multimodal quality checks, and diagnostic risk alerts. Deployed and in use at three hospitals, with positive results.
From a One-Sentence Request to an AI-Native Service Desk for a Healthcare Group
Seventeen days from a three-agent concept to real production. I came away with a different understanding of AI-native enterprise software—not a model taking over everything, but a closed loop across natural language, business governance, reliable transactions, and organizational learning.
Intelligent Clinical Documentation and Quality Control
Embed AI in the clinician’s real workspace so generation, quality control, and sign-off reduce repetitive documentation while turning validated practice into reusable institutional knowledge.
Patient Agent and Precision Patient Operations
Move beyond digital entry points toward a continuously available health twin that understands patient state, initiates action, and completes clinical service loops under human confirmation and full audit.