[ CASE STUDY / 02 ] · DENTAL ALIGNERS AND OPERATIONS PLATFORM
MySmartalign
A multi-tenant platform for dental clear-aligner orders, 3D treatment tracking and AI-assisted customer support.
- Role
- Full Stack Developer
- Company
- Akkenam Technologies
- Period
- May 2023 – Dec 2024
- Stack
- Node.js · Express.js · GraphQL Yoga · MongoDB
This is company work, so the source code and screens stay private. This page explains the engineering instead.
[ 01 ] · THE PROBLEM
Aligner orders involve a lot of moving parts. The platform had to keep every organisation’s data separate, make a treatment easy to picture, and support customers with routine questions about their orders.
[ 02 ] · HOW IT FITS TOGETHER
Architecture
The React client with its Three.js tooth model calls one multi-tenant GraphQL Yoga API, which uses Redis, PostgreSQL, MongoDB and the OpenAI API.
[ 03 ] · WHAT I DECIDED
Key decisions
01
A structured GraphQL API for every tenant
One GraphQL Yoga API serves every organisation on the multi-tenant platform, with a clear schema for orders, treatments and support.
02
Transactional and flexible storage side by side
Order records that need strict consistency live in PostgreSQL. Data that changes shape lives in MongoDB. Each store does the job it is good at.
03
Treatment you can turn around
An interactive 3D tooth model built with Three.js lets a clinician inspect the treatment visually instead of reading a table.
04
Cache the things every request needs
Redis holds tenant metadata and sessions, so the common lookups do not hit the database on every call.
05
AI where it removes a wait
AI-assisted order and delivery support, built on the OpenAI API, handles routine questions about an order.
[ 04 ] · WHAT CAME OUT OF IT
Results
- Orders, 3D treatment tracking and customer support in one multi-tenant platform.
- Tenant metadata and sessions served from Redis instead of the database.
- A 3D view clinicians can inspect interactively, not just a static report.
ALL CASE STUDIES