I've been building a local AI assistant called Arwanos that runs
100% on your own machine — no cloud, no API keys, no data leaving
your device.
The latest version (v10) adds something I'm genuinely proud of:
a Mental State Monitor — an ML pipeline that reads your personal
journal, cross-references your patterns against 7,557 real therapy
session examples, and generates psychological questions that get
deeper with every session.
A few things it does that I think are interesting:
- Builds a psychological profile from your journal history before asking anything
- Questions never repeat — a 3-layer NLP check prevents it
- Runs entirely offline after a one-time dataset build
- Depth scales with session count (early sessions = surface, later = deep)
I wrote a full technical breakdown of the ML pipeline on my site —
phases, dataset sources, anti-duplication logic, and how the insight
pipeline works after each session.
Source code: 
Questions or thoughts? Drop a comment — happy to discuss.
Top comments (0)