Your notes, in your voice.

NoteWright sits downstream of your AI scribe and rewrites its output to match how you actually document — your structure, your phrasing, your clinical reasoning density. Not a template. A learned representation of your style.

Built by a practicing family physician for his own charts. Now preparing for release.

The problem

AI scribes transcribe the encounter well enough. Then they hand you a note that reads like it was written by someone else — because it was. Generic structure, foreign phrasing, reasoning that's either padded or missing. The editing tax eats the time the scribe saved.

What NoteWright does

Paste your scribe's output. Pick the note type. Get it back written the way you write — then finish it on an editing surface built for the job: resolve placeholders, accept or reject suggested upgrades, and paste a clean note into your EHR.

Principles

1. Nothing is ever invented. If a value isn't in your input, it isn't in your note. Anything unresolved stays a *** — exactly where your eye, and your EHR, expect it.
2. You compose; it assists. Suggestions are transparent and yours to reject. No silent inference, ever.
3. Your style belongs to you. NoteWright learns your documentation voice from your own notes and feedback — it doesn't impose a house style.

Where this goes

Today, NoteWright learns your style through a structured profile and your ongoing feedback. The endgame is a personal model — fine-tuned on your own notes, running on your own hardware. Your voice, learned by a model you own, on a machine you control. That destination is why NoteWright is built the way it is: outside the EHR, downstream of the scribe, accumulating the one dataset nobody else has — how you write.