The AI scribe helping 2.7 million patient visits a week, and why its database matters as much as its model

Key points

• Heidi's AI scribe product supports roughly 2.7 million patient interactions per week across more than 190 countries, as of 2025.
• A pilot at Beth Israel Lahey Health, one of New England's largest hospital networks, found 74% of clinicians reported less after-hours paperwork after using Heidi.
• Heidi stores all patient data inside the same region where care was delivered, enforced by infrastructure rather than policy promises.
• Migrating to MongoDB Atlas, a cloud-hosted database service, cut response times on key parts of the app by nearly 33%, according to the company.
• Heidi co-founder and CTO Yu Liu says the AI model itself accounts for roughly 20% of what makes the system work; data architecture carries the other 80%.

A doctor in London finishes a consultation. Within seconds, a structured clinical note appears. No dictation. No typing. The doctor reviews it, makes a change or two, and moves on to the next patient.

That is what Heidi Scribe does, and it is doing it roughly 2.7 million times a week.

Heidi is an Australian company that builds AI tools for clinicians. Its flagship product listens to a clinical consultation and automatically drafts the documentation that follows, the notes, referrals, and structured records that can consume hours of a doctor's day. The company now operates in more than 190 countries.

What makes healthcare AI different from every other kind?

A two-percent error rate in most software is a minor nuisance. In a clinical setting, it is a patient safety problem.

"The architecture has to be built around the assumption that every output may be scrutinised, audited, and relied upon in a patient's care," says Yu Liu, Heidi's co-founder and chief technology officer.

That shapes every engineering decision Heidi makes. Data residency, meaning the rule that a patient's records must stay physically within the country or region where care happened, is not optional in healthcare. A clinician in Sydney works under Australian Privacy Principles. One in London works under GDPR, the European Union's data protection law. One in Tokyo falls under Japan's APPI. One in Denver falls under HIPAA, the US health data law. Heidi runs a fully separate, isolated deployment in each region to satisfy all of them at once.

Auditability is equally non-negotiable. If a regulator or hospital asks what the AI saw during a session three months ago, what it produced, and what the clinician changed, Heidi must be able to answer. That capability has to be designed in from the start, not bolted on later.

Why the database choice was a founding decision

A single Heidi Scribe session is not one tidy file. It is a collection of transcripts, structured notes, templates, patient context, records from electronic health systems, and dozens of other pieces that all belong together and all change shape as the product evolves.

Traditional databases store data in rigid rows and columns, like a spreadsheet. That structure breaks whenever the data model needs to change, which in a fast-moving AI product is constantly. Heidi chose MongoDB Atlas, a cloud database service that stores data as flexible documents rather than fixed rows, letting the team reshape records without freezing the whole system for a migration.

"The model is maybe 20% of the system, and the data architecture is what determines whether the other 80% holds up under real clinical load," Liu says.

The move paid off in speed, too. Switching to MongoDB Atlas cut latency on key APIs by nearly 33%, the company says.

How the AI actually retrieves clinical knowledge

Heidi uses a technique called RAG, short for retrieval-augmented generation. Instead of asking an AI model to recall medical facts from training data (where it might confidently produce something wrong), RAG pulls verified, up-to-date information from trusted sources and hands it to the model as context.

Heidi's sources include licensed clinical knowledge bases from organisations like BMJ Best Practice and NICE CKS, a UK clinical guidance body. The system is jurisdiction-aware: a UK clinician gets UK treatment guidelines; an Australian clinician gets Australian drug formularies. The right answer in one country can be the wrong answer in another.

Citations are hard-coded to their source records, not generated by the model. The AI cannot invent a reference.

What does this mean for patients and clinicians?

For clinicians, less time on paperwork means more time with patients, or simply finishing at a reasonable hour. The Beth Israel Lahey Health pilot captured that directly: 74% of participating clinicians said after-hours documentation fell after using Heidi.

For patients, the promise is a doctor who is present in the room rather than half-focused on a screen. The risk, as with any AI clinical tool, is over-reliance on outputs that have not been carefully reviewed. Heidi's design keeps the clinician as the final check: every note is reviewed before it becomes part of the record.

Heidi is now building toward a fuller view of the clinical workflow, covering everything from pre-visit preparation to post-visit referrals and follow-up automation.

Common questions

Is Heidi approved as a medical device?

Heidi markets its product as a documentation assistant, not a diagnostic tool. Regulatory classifications vary by country; clinicians remain responsible for reviewing and approving every note the system produces.

Can a hospital trust that patient data stays in their country?

Heidi says yes, and says it is enforced by infrastructure: each country or region runs its own fully isolated database cluster. The data physically cannot cross a regional boundary, rather than being kept in place only by a contractual promise.

Does using an AI scribe mean the AI is making clinical decisions?

No. Heidi Scribe records and drafts documentation. The clinician reads it, edits it, and signs off. The AI does not diagnose or prescribe.