I spent over a decade as a Radiologic Technologist. Radiology, MRI, CT, emergency departments. COVID ER. The kind of work where the infrastructure you depend on is invisible until it fails — and when it fails, people notice immediately.
PACS systems, HL7 interface engines, RIS workflows — none of this is glamorous technology. It runs in the background of every scan, every diagnosis, every treatment plan. When it works, nobody thinks about it. When it breaks at 2am, a radiologist is on the phone and a patient is waiting.
That is where I learned what reliability actually means. Not as a concept. As a consequence.
The move
The transition to cloud security engineering was not a pivot away from healthcare. It was a move toward the infrastructure layer that healthcare depends on but rarely thinks about. The same instincts apply: what breaks when this fails, who notices first, and how do we know before they do.
In radiology, we ran redundant PACS servers, tested failover on nights nobody asked us to, and kept paper processes ready for when everything else failed. Not because we were told to. Because we understood the stakes.
Now I build Lambda functions with dead letter queues and CloudWatch alarms. The vocabulary changed. The discipline did not.
What transfers
The clinical background gives me something most cloud engineers do not have: a firsthand understanding of what the data actually represents. When I write IAM policies that protect patient data in S3, I know what that data is. I have handled it. I have protected it in physical spaces before I protected it in logical ones.
HIPAA compliance is not abstract to me. It is the difference between a patient imaging study staying private and it not. That changes how you approach a security control — not as a checkbox but as a boundary that matters.
Building in public
chamb.dev is where I document what I build. The Cloud Resume Challenge led to a compliance engineering capstone, which led to a personal brand platform, which led to the system that published this post automatically while I was doing something else.
Everything here is real. No tutorials. No demos. Production infrastructure with real consequences — and the same operational mindset I carried out of the ER.