Pre-launch draft. Demo encounters are synthetic and have not yet been reviewed by a clinician; accuracy figures are not yet published.

P2 · Documentation

Bolo Scribe

Ambient documentation of a code-mixed encounter.

Book a pilot call

Endocrinology follow-up · 58F · Gujarati + English

Not clinically reviewed

No medication change is ever written to the chart by the system. Draft only; the clinician signs.

Synthetic demo encounter. Not a real patient.

2

jobs at once — listening, and translating while listening

Your physicians finish clinic and start documentation. On a panel that speaks Gujarati, Telugu or Hindi at home, they are also doing a second job during the visit: translating in their head while they listen. The ambient scribes on the market help with the first problem and make the second one worse, because they transcribe the English and lose the rest.

A physician listens to an older patient's chest with a stethoscope in a clinic examination room.
A physician listens to an older patient's chest with a stethoscope in a clinic examination room.

How it works

  1. 01

    Pull context before the visit — active medications, problems and allergies become a bias set for the second pass.

  2. 02

    Capture ambiently, decode the code-switching, and separate clinician from patient from companion.

  3. 03

    Link clinical entities and score every quantity separately from the text around it.

  4. 04

    Compose a draft note where every assertion traces to a transcript span — and an after-visit summary in the patient's language.

What it will not do

Stated plainly, because a limit you find out about later is a limit you stop trusting the rest of the page over.

  • It does not write to the chart. The note is a draft until a clinician signs it, and there is no configuration flag that changes this.
  • It never commits a medication change. Dosages below the confidence threshold render amber and must be touched by a human.
  • It does not diagnose, suggest a diagnosis, or flag a condition. It documents what was said.
  • No entity appears in the note that does not trace to a transcript span with a timestamp and a confidence score. If it cannot be traced, it does not ship.

Integration

Every buyer asks within ninety seconds of the demo starting. A plain FHIR path exists today — “here is the bundle” is a real answer, and “we are in the marketplace review process” is not.

FHIR R4 DocumentReference
Draft status until signed. Available today.
FHIR MedicationStatement, Condition, AllergyIntolerance
Read, to build the pre-encounter bias set.
HL7 v2 ADT / SIU
Still the reality at most small practices.
Epic App Orchard · athenahealth Marketplace
In progress. The FHIR bundle does not wait for it.

What we measure

These are the numbers for this product. None of them are published yet — see the methodology for why an unstratified figure would be worse than silence.

Time-to-signature
What the clinician actually feels. The headline product metric, not WER.
Note acceptance rate
Percentage signed with no substantive edit.
Entity F1 by code-mix band
The number that predicts clinical harm.
Dosage exact-match
Patient-safety metric. Reported on its own.

What you are probably about to ask

We already looked at two ambient scribes. Why would this be different?

It probably is not, for your English-speaking panel — they are good products. The difference shows up on the encounters where your patient starts a sentence in Gujarati and finishes it with a drug name. Ask any vendor for their accuracy on utterances that are 30–60% English and see whether they can produce the number at all.

What happens when it gets a dosage wrong?

It renders amber, labelled low confidence, and it is never auto-committed. There is no configuration flag that turns that off. Every quantity is scored separately from the text around it, precisely because that is the error that reaches a patient.

How does it get into our EHR?

FHIR R4 DocumentReference, in draft status until your clinician signs. A plain FHIR export path exists today; marketplace listings take months, and we would rather you had a working bundle next week.

Pilot

Book a pilot call.

Thirty minutes. We will run the demo on the languages your panel actually speaks, show you the eval methodology, and tell you plainly which parts are not ready. If we are not a fit we would rather find out on this call.

One reply from a person. We do not run a nurture sequence and we do not share your address.