Body
Training, physiotherapy, health markers, nutrition, and the niggles that turn into injuries if nobody is counting.
sessions and load · pain scores · recovery markers · sleep · body composition · nutrition targets
We build systems that pay attention.
Resonance Labs is a New Zealand venture studio. We make software that runs on your own data, keeps it on your own hardware, and tells you plainly what it found.
- Studio
- Resonance Labs Limited
- Based
- Aotearoa New Zealand
- Product
- Sol, for iOS
- Status
- In build
What we build
Most personal software asks you to feed it. You log, it stores, and the value comes back as a chart you are left to interpret on your own. The work moves to you, and the software takes the credit.
We build the other shape. Systems that read what you already produce, compute everything that can be computed, and reserve judgement for the few places judgement is actually required.
The distinction
A tool shows you your data. A system decides what to do about it, says why, and shows its working so you can disagree.
Sol
Sol is System Operating Life: an iOS app and the engine behind it. It plans each day, tracks goals against real baselines, and reports, all from data you already generate.
It runs two pillars, Body and Mind, and treats them as one system. Training load reads against sleep and recovery. Mood reads against load and life events. Neither is interpreted in isolation, because neither happens in isolation.
Not a diagnosis
Sol observes and flags patterns. It does not diagnose. When a marker regresses or a decline holds, it says so and points to the clinician who should see it. It informs the people who treat you; it does not replace them.
How it works
Five layers, each with one job. The separation is the design: it is what stops a language model from quietly inventing a number.
| Human truth | Markdown files you can open, read and edit No database you cannot inspect. |
|---|---|
| Machine truth | An append-only event log, written once, never rewritten Bi-temporal: when a fact was true, and when it was learned. |
| Computation | Scripts calculate volumes, trends, progress and targets Every number is produced by code that can be re-run. |
| Interpretation | The model reads the results and explains them It never does the arithmetic itself. |
| Delivery | The plan, the summaries and the goal data land in the app Plus a heartbeat, so a failed run is visible rather than silent. |
Two pillars
Each pillar owns its own metrics and its own rules. Adding a third would not change the platform, which is the point.
Mind
Mood, energy and stress over time, habits that hold or slip, and the life events that explain a week the numbers alone cannot.
mood · energy · stress · habit streaks · life events · cross-pillar patterns
What we hold to
-
Your data stays on your hardware.
Inference is the only thing that leaves. Everything at rest stays on machines you control.
-
Scripts compute, models interpret.
No metric arithmetic in prose, ever. If a number appears, code produced it and you can re-run that code.
-
Markers gate progression, not the calendar.
A plan advances when recovery and health markers say it can, not because a week ended.
-
You are the authority.
Your override wins immediately, is logged in your own words, and is never quietly argued with.
-
Silence never means success.
Every run stamps a heartbeat. A failure says it failed. Nothing is inferred from the absence of a message.
-
Nothing is deleted.
Superseded data is archived, not removed. The record of what you believed last year survives changing your mind.
Why write these down
These are constraints on us, not features for you. Each one costs something to keep, which is the only reason a list like this is worth reading.