About
One method, two materials. Medicine and engineering are not adjacent careers. They are the same discipline applied to different kinds of evidence.
- Published
- Tags
- Profile
A sentence can sound finished and still be wrong. That is the problem I work on. In clinic, in software, and in the systems that sit between them, the failure mode is the same: a fluent answer with no provenance, offered as if fluency were the same thing as knowledge.
I treat medicine and engineering as one method. Enter a domain where information is fragmented or easy to dress up. Make the source of a claim explicit. Keep uncertainty visible instead of smoothing it into confidence. Turn the structure into something a person can actually use. Then measure whether it still holds when it leaves the machine, or the notebook, that produced it.
Path
I moved to the United States in 2015 and completed a B.S. in Biology at the University of Colorado Denver in 2019. The years after that were product, data, and an early medical-AI venture. I began medical training in 2022. I am in clinical training now. The engineering is not a side project stacked on top of that. It is the other half of the same work.
SkillMap is how a skill library becomes a governed registry. Medical Study OS keeps source, notes, and retrieval in one inspectable workspace. Persistent memory is scoped retention with a hard privacy boundary. XOT is live production, not a model call. THI.AI and Sabz Ventures are earlier. 0x3.dev is the umbrella for selected external work.
How to read this
The list is compact on purpose. Click through for the case. Status labels are defined, with the rest of the publication rules, on Claims and boundaries.
If you want the operating discipline, read How I Build. If you want the current desk, read Now.
Public contact is through 0x3.dev. Private email and phone are not published here.