Your life, their servers
Most AI assistants send everything you say and store to remote servers. For a genuinely personal assistant, that trade is the product's original sin.
An on-device personal AI assistant app: the intelligence lives on the phone itself, so personal data never has to leave it.
On-device by design
Most AI assistants send everything you say and store to remote servers. For a genuinely personal assistant, that trade is the product's original sin.
Cloud assistants stall exactly when connectivity does, which in much of the world is often.
Assistants built for everyone remember no one. A personal assistant should be shaped around one person's work and habits.
AURA runs as a native-feel Flutter application on the phone.
Requests are processed by the on-device model, not shipped to a server.
Personal information remains in the phone's own storage.
AURA is my own design and build: the product concept, the on-device architecture choice, and the Flutter application, including its prototype layer.
It is also a research vehicle: what I learn here about small, local models feeds the wider NeuroParsec Lab work on private AI.
Original build by Simanto Mitro.
The Flutter build and prototype are in progress locally. There is no public release and no repository yet, and this page does not claim one.
If this kind of system would help your clinic, organisation or team, the conversation starts with one cup of coffee.