Two photos of one forearm scored badly. The lighting hadn't changed. The background had, and it filled most of the frame. The app was measuring the room. Cropping to the subject halved the distance.
Independent software studio · Albuquerque
Ramirez Labs builds software for iPhone, Mac, and Vision Pro. The work runs on-device by default, no accounts, no servers, nothing uploaded, because the things worth tracking closely are usually the things you'd rather not hand to somebody else.
Measurements taken during development. Each one changed a decision.
Two photos of one forearm scored badly. The lighting hadn't changed. The background had, and it filled most of the frame. The app was measuring the room. Cropping to the subject halved the distance.
A well-framed patch of skin scored lower than an out-of-focus keyboard. The metric was reading surface texture, not sharpness. No threshold fixes a signal that measures the wrong thing, so it was replaced with the gyroscope.
Apple's on-device language model mapped speech to structured actions correctly 76% of the time. A deterministic validator around it caught most of the rest, without touching the model.
A photo journal for watching one spot on your skin over time. Your last photo appears faintly over the camera so you can line up the same shot, and each new picture is compared against your first using Apple's Vision framework, entirely on the phone. It reports how much has changed, never whether that's good or bad. That judgment isn't the app's to make.
Point your camera at a spider, a flower, a bird, and find out what it is. One scanner instead of the four apps this usually takes, with a bias toward the question people actually have first: is this one dangerous? Named for the sampling frame ecologists lay on the ground to survey a plot.
Patient information placed around a clinician instead of on a screen they have to turn away to read. Voice dictation sorts what's said into the right part of the note. Built to explore whether spatial computing does anything real for the documentation load that drives clinician burnout.
A voice assistant running on Apple's on-device model, with no network involved. Alongside it, a test harness that runs a fixed set of sentences through the model repeatedly and reports how reliably speech becomes a structured action. That's how the 76% above got measured, and how the validator that lifted it got built.
Ramirez Labs is Alexander Ramirez. By day I write software for a state health agency; the rest of the time I build the apps above.
Working inside healthcare shapes what I make. Systems that handle personal information tend to collect more of it than they need, so these apps collect none, no account, no cloud, no analytics. It's a constraint worth designing around rather than apologizing for.
I'm also a doctoral student in computer science, looking at how well small on-device language models turn ordinary speech into reliable actions. The measurements above come out of that work.
Currently open to iOS, macOS, and visionOS work, especially in healthcare, simulation, and training. Relocating to Orlando.