All work · Healthcare
Five acquired healthcare products, nine programming languages, one gRPC data access layer, all migrating onto Azure under HIPAA, HITECH, HITRUST and SOC 2 without a big-bang cutover.
A healthcare company had grown by acquisition and owned separate codebases, each with its own stack and one to five databases of its own. They needed one coherent platform, in a regulated setting, and they could not stop serving customers while they got there.
I own the target architecture, the data models against SNOMED, LOINC and RxNorm, the API contracts, the migration scripts, and the CI/CD pipelines with security and compliance gates built in, along with the Azure service selection, network design, tenancy model, least-privilege identity and DR against approved RPO/RTO targets. Features move one at a time out of legacy, legacy and unified run side by side, and each legacy system is decommissioned only once its replacement is verified in production. The interop surface is SMART on FHIR, HL7 FHIR and v2, X12 EDI, C-CDA, DICOM, XDS/PIX/PDQ, Kafka and event hubs, webhooks and secure FTP.
The unifying data access layer is built, working and tested. It spans nine programming languages, reaches every acquired product's stack and databases, and pulls all data-access code into one place, so the refactor happens once rather than in twenty places. It runs over gRPC, against internal pushback that a common gRPC layer would be too slow. I have final technical decision authority within approved scope and the go/no-go on every cutover.
Stack
Tell me what it does, what it is written in, and what breaks when it goes down. I will say plainly whether I can help.
Tell me what it runs on