How we work

Approach.

A practice built on three traditions, held in tension with one another.

02 · Thesis
The work

Most data architectures are inherited, not designed.

They accumulate, piece by piece, decision by decision, until no one remembers why the system looks the way it does. The original architect leaves. The team turns over. The documentation drifts out of sync with the code that produced it.

What Mojima practices is the careful, intentional construction of architectures that hold up when the people who built them are gone. That requires three kinds of work, and they have to be done together.

We call them Heritage, Precision, and Intelligence.

03 · Heritage

HeritageMokoto

We begin by understanding what was already here.

i

Listen before designing

Every engagement begins with a period of careful listening, to the systems, to the people who built them, to the constraints that aren't written down. Architecture without context is decoration. We do not propose solutions in the first week.

ii

Map the decisions, not just the code

Systems are made by people making decisions under pressure. Understanding why something is the way it is tells you whether to keep it, replace it, or work around it. We document the why, not just the what.

iii

Honour what works

Not everything needs replacing. Some systems that look dated are simply mature: they survived because they handled the real problems well. We are careful not to rebuild for the sake of rebuilding.

04 · Precision

PrecisionJinkou

We design with care for what comes next.

i

Plain language documentation

We write things down, designs, decisions, trade-offs, in language a non-specialist can follow. Documentation that requires the original architect to interpret it is not documentation.

ii

Built for handover

The test of a good architecture is whether it can be maintained by the team that inherits it. We design with the inheritors in mind, not with our own elegance in mind. Boring solutions outlast clever ones.

iii

Resist cleverness

Every system carries a temptation to demonstrate sophistication. We name those temptations explicitly and decline them, in favour of solutions that are obvious, predictable, and traceable. Sophistication is a liability when the original author leaves.

05 · Intelligence

IntelligenceMātauranga

Bespoke is judgement, not assembly.

i

We do not productise

There is no Mojima methodology to buy. Each engagement asks a different question, and each one earns a different answer. We bring experience and discipline; we do not bring templates.

ii

We say no often

Not every engagement is the right fit, and not every problem is one we should take on. We turn down work where the constraints make good architecture impossible, or where the timeline forces shortcuts we cannot defend.

iii

We adapt the practice to the work

Some engagements need a single architect for six weeks. Others need a small team for two years. We do not force the shape of the work to fit the shape of the practice. The work decides.

06 · In practice
Where it leads

These do not work in sequence.

They are not phases. Every engagement, no matter its size or shape, holds all three at once. That is what bespoke actually means.