Designed for the people it serves
For an ADHD and autism community, the difference between a useful app and an abandoned one is friction, not features. Every password is a barrier; every cluttered screen is a cost.
So the brief inverts the usual order. Clarity, low cognitive load, and reminders that actually fire are the product. Everything else is negotiable; accessibility gives way to nothing.
Two things, done properly
Events. People see the sessions they have already registered for, get customisable reminders before each one, and join the call in a single tap. The app reads from the systems the organisation already uses; it does not become a second registration system, and it does not collect what those systems already hold.
Resources. The organisation's support content, rebuilt for low cognitive load: a small number of clear, guided paths to the right resource rather than a search box that puts the burden of knowing what to ask on the person who needs help.
No sign-in is needed to browse either. Verification only appears at the one gate where it earns its place, seeing your own events, and after that a fingerprint or glance gets you back in.
The shape of the build
- Platforms Android and iPhone from day one, built once on a shared codebase.
- Identity One-time email code plus device biometrics. No passwords, no accounts to manage.
- Data Minimisation by architecture: personal data stays in the systems that already govern it. The app holds no warehouse.
- Standard ICO Children's Code applied to every user, and a Data Protection Impact Assessment completed before launch.
AI, on a leash by design
A second phase adds an AI resource helper: ask in plain language, it finds the right existing resource. Two hard rules: retrieval-only, never generating advice of its own, and a mandatory crisis-escalation path designed in from the first line.
It also ships only after the launched app has run in the real world long enough to validate it is safe. That order is deliberate. For an audience like this one, an AI feature that is probably fine is not fine.
The same discipline behind every build
This is not what we usually build; it is here because it shows how we build. Data stays in the client's systems, safeguards are designed in, and AI gets exactly as much freedom as the context can safely bear. If your business handles data that matters, this is the discipline your build gets.