Minikai connects to the data your organisation already has — whether that's in Lumary, Salesforce, Nightingale, SharePoint, or uploaded directly. That data gets synced or imported into Minikai, where each person's records are indexed and made searchable by their Mini.
When you ask a Mini a question, it searches across the records on that person's timeline — progress notes, care plans, assessments, incident reports, charts, attached documents — and then generates a response grounded in what it finds. Every response includes numbered citations so you can see exactly which records the information came from.
Minikai is powered by a large language model that's been configured for the care sector. It understands NDIS and aged care terminology, uses person-centred language, and knows how to structure common documents like progress notes, family updates, and progress reports.
ℹ️ Good to know: A Mini doesn't browse the internet or pull information from external sources. It only works with the data that's been provided for that specific person. All AI processing happens in Australian Azure data centres.
What a Mini produces is always a draft or a summary — not a clinical decision. You review, edit, and apply your own judgement before using anything it generates. For more on this, see Human in the Loop.
💡 Tip: You can add records to a Mini's knowledge by importing documents or importing from a spreadsheet. The more current the data, the better the responses.
Need help?
If you have questions or something isn't working as expected, message us through the help icon in the bottom-right corner of Minikai. We're here to help.