What it is
A self-hosted system for preserving raw wearable measurements and turning them into explainable daily features, longitudinal analysis and personal experiments. Python calculates the statistics; AI explains structured results.
How it works
A recovery score should come with its inputs, missingness and algorithm version. Somatriq keeps the raw evidence so a change in interpretation does not erase what the device actually measured.
Status: active prototype. Server, analytics and web features are implemented; mobile sync lifecycle, hardware acceptance and deployment validation remain open. Dashboard screenshots use synthetic data. Somatriq is independent of WHOOP and is not a medical device.
Results
| Preserve | Understand | Experiment |
|---|---|---|
| Immutable raw frames and canonical observations | Recovery, HRV, sleep and longitudinal context | Behavior journals and N-of-1 experiments |
| Idempotent ingestion and export | Visible coverage and versioned calculations | Deterministic statistics with uncertainty |
Try it
git clone https://github.com/Gjusev/somatriq.git
cd somatriq
uv sync
uv run pytest -q
cd apps/web
npm ci
npm run typecheck
npm run build

