Mariete Edge Foundation
Edge is on-device AI. Inference runs entirely on the phone using on-device Gemma; storage is encrypted at rest with biometric lock; nothing leaves the device after first-run setup. Spaces partition context so that one client matter never bleeds into another. This Foundation course teaches the operator side: how to set up Spaces correctly, how to brief work that has to stay confidential, how to read the output knowing it is local, and how to adopt the hygiene routines that keep the device defensible as a working surface for privileged material.
What you'll be able to do.
- Explain edge is in your own words.
- Setting up Spaces
- Reading Edge output
- Recognise the four common workflows.
4 modules. ~60 minutes total.
- 01What Edge is~14 min
Edge is not a cloud assistant on a phone. It is not an offline cache of a server model. Edge is an on-device AI — model weights, inference,…
- 02Setting up Spaces~15 min
A Space is the operator's primary tool in Edge. A well-configured set of Spaces makes confidential work effortless and correct. A messy set…
- 03Reading Edge output~15 min
Edge is on-device, which means the output you read has properties — and limits — that cloud assistants do not have. Operators new to Edge of…
- 04Common workflows and hygiene~16 min
Edge is most valuable when it is woven into operator routines — not opened occasionally, but used as the default working surface for confide…
One sample question. Click to reveal the answer.
Which statement best describes what Edge is, according to this course?
- A thin-client app that streams prompts to a Mariete cloud model and renders responses locally
- An on-device AI where model weights, inference, storage and memory all live on the phone, locked behind biometric
- An offline cache of a server-hosted assistant that syncs when signal returns
- A multi-device assistant whose context syncs across phone, laptop and desktop
15 questions like this. Edge Foundation pass-mark is 75%.