Healthcare & Disability ServicesPhase 1 delivered
Movido Care
AI Support Plans for a Disability Care Provider
Intake forms and handwritten notes turned into a structured draft support plan, with consent, masking, encryption and an audit trail built in.
10
Plan sections drafted
Masked
NDIS number
3
Access roles
AES-256
Field encryption
The problem
New participants complete a 34-page intake form. Someone then had to turn that information into a support plan by hand. The information is sensitive health and disability data, so a general-purpose chatbot was never an option.
What we built
- An intake wizard, plus upload of PDFs and photos of handwritten notes, with AI extraction into structured fields.
- Draft generation of a 10-section support plan, streamed as it writes, and exported to Word for staff to review and finish.
- A participant list and an admin audit page.
Privacy and safety built in
- A consent step: staff confirm the participant has agreed, see where the data will be processed, and each consent is recorded in the audit log.
- The NDIS number is masked by default; revealing it is role-checked and logged.
- Three access roles, checked on every page.
- Intake data and the NDIS number encrypted at rest with AES-256-GCM.
- An audit log that the application only ever appends to.
- Secure sessions, hashed passwords and rate-limited sign-in.
- Every plan checked against a fixed structure before it is exported.
What comes next
The system supports the provider's compliance work rather than replacing it. The hardening roadmap covers multi-factor sign-in, an audit trail enforced by the database and kept for seven years, a data protection impact assessment, and processing within Australia.
Technology
- Next.js
- TypeScript
- Anthropic Claude
- Postgres
- Vercel
- Word (DOCX) export