Tooth Fairy: Building the UK's First Health-Regulated Dentist On-Demand App
Ethereal Softech has served as the engineering team behind Tooth Fairy since 2021, delivering the native apps, the microservices backend, and the practitioner dashboard described below. The platform shipped in 2022, and new features and enhancements are still in active development with us today.
The Challenge
The UK's dental system had a problem you could measure before the pandemic and feel after it. £18 million a year was already being spent on A&E attendances for dental pain, with people turning up to emergency rooms because they couldn't get a dentist appointment. Only 3 in 10 children visited a dentist when they were in pain (NHS 2016/17). Then COVID arrived and 19 million dental appointments were missed across the UK, accelerating a postcode lottery that dentists on the ground had been describing for years. Tooth Fairy was started by practising dentists who had treated thousands of patients and watched them fall through those gaps. They weren't interested in building another appointment booking tool. They wanted to create the world's first dentist on-demand: a platform where any UK patient could video call a qualified dentist today, not in three months, for a live consultation, an AI-assisted treatment review, an emergency triage, or a cosmetic assessment. The bar was deliberately high: the platform had to be tested and validated in the NHS, become the UK's first health-regulated dental app, and serve both individual patients and large enterprise clients like NHS trusts and FTSE-listed companies on the same infrastructure.
Our Approach
The first call we made was also the most deliberate one: native apps, not cross-platform. Kotlin for Android, Swift for iOS. We know React Native and Flutter ship faster. We chose against them anyway. In a health-regulated dental app, the camera is not a feature. It is the consultation. When a patient holds their phone up during a live video call to show a dentist a cracked tooth or an inflamed gum, image quality, focus response, and colour accuracy directly affect what the dentist can see and diagnose. The AI-enabled treatment tracking feature analyses patient dental photos, which means image fidelity is diagnostic fidelity. Cross-platform frameworks abstract the camera layer. For a product validated in the NHS, we needed the hardware control that only native gives you.
The backend is not a single server. It is a suite of Node.js microservices, every one of them written in TypeScript. Separate services own separate domains: appointment scheduling, dentist availability, video session management, user accounts, corporate account administration, and AI treatment tracking. They communicate internally via gRPC (Protocol Buffers over HTTP/2) rather than REST. For a platform where booking confirmation, practitioner availability checks, and video session setup all happen in sequence in real time, gRPC's speed and strongly-typed service contracts matter. TypeScript enforcing those contracts on both ends catches integration errors at build time, not in production with a patient waiting on screen.
Each microservice runs in a Docker container. That means individual services can scale independently. When a Telegraph feature drives a booking surge, the appointment service scales without touching video or user services, and deployments stay consistent across environments without manual configuration drift. MongoDB stores the primary data: patient records, appointment history, consultation notes, treatment tracking data, and corporate account structures. Document storage is the right fit here because dental consultation records are not uniform. A teeth-straightening treatment journal looks nothing like an emergency triage note or a cosmetic consultation summary. MongoDB handles that variability naturally.
Redis sits above the database as the session and caching layer. During peak windows, it keeps authenticated request response times fast without routing every call back to MongoDB. It also handles the session invalidation requirements that come with storing patient data on a health-regulated platform. React.js drives the practitioner web dashboard, the desktop interface where dentists manage their schedule, review patient histories, and join live video sessions. Every dentist on the platform is specifically trained for remote consultation before their first session. The same dashboard provides a separate enterprise layer for corporate accounts, managing dental plan entitlements for employees at NHS trusts, HM Government departments, Rolls-Royce, DHL, Sainsbury's, Unum, and DEFRA.
What We Delivered
- Native iOS (Swift) and Android (Kotlin) patient apps, built for diagnostic-grade camera fidelity in live video consultations
- A TypeScript microservices backend (appointments, availability, video sessions, accounts, corporate administration, AI treatment tracking) communicating over gRPC
- React practitioner dashboard with a separate enterprise layer for corporate dental plans
- Dockerised, independently scalable services with a Redis session/caching layer meeting health-regulated session-invalidation requirements
Where the Product Is Today
The platform this engineering underpins has gone on to:
- Proprietary technology patented in 22 countries
- Recognition as the UK's first health-regulated dental app, tested and validated in the NHS
- App of the Year and Digital Health Solution of the Year; Highly Commended, High Tech Launch of the Year
- 10,000+ Google Play downloads and a 5-star Trustpilot rating (as of July 2026)
- Enterprise dental coverage across NHS trusts, HM Government, Rolls-Royce, DHL, Sainsbury's, Unum, and DEFRA
- Press coverage: ITV, BBC, Sky News, The Telegraph, Forbes, TechCrunch, Vogue, and Vanity Fair
Technologies Used
See It Live
More Projects
Want Similar Results?
Book a free 30-minute call. We'll explain exactly what we can achieve for your business.


