Instrument room
What each instrument reads.
The object changes. The discipline does not: infer hidden structure from observable traces, expose the evidence, state the limit.
Facet
[ LIVE ]- It reads
- The multidimensional shape of a codebase.
- It returns
- A fourteen-axis profile with line-cited findings and concrete fixes.
- It is for
- Teams deciding where software quality risk actually sits.
ConstructLens
- It reads
- Latent constructs expressed through behaviour over time.
- It returns
- Evidence-linked construct scores with reliability and coverage limits.
- It is for
- Researchers and teams piloting defensible agent or human measurement.
Aletheia
- It reads
- Measurement error hiding inside scientific conclusions.
- It returns
- An audit of what changes when uncertainty in the measure is modelled.
- It is for
- Research teams whose conclusions depend on noisy measures.
Custom measurement and audits
- The work
- Not every question fits an instrument off the shelf. I build bespoke measures for a quality that is specific to your setting, and I audit existing ones, the cases where a number that looks solid turns out to be measuring something else. If you have a measurement problem that actually matters, start a conversation.
Client zero
The first user is the builder.
I am the first user of everything I build. Each instrument runs on my own work before it goes near anyone else's. Facet read my code long before it read yours, and that is most of why I trust what it tells you.
My training is in measurement, not software engineering, and I am not going to pretend otherwise. Lemma is the learning surface I built to close that gap, and I work through it most days. It points at something I keep coming back to: the way a person builds capability can be measured and developed with the same discipline as anything else here. That is most likely where the work goes next.