Northing Vision

I’ve spent roughly half my career delivering civil infrastructure projects and half building software for the people who deliver them. They might seem like different jobs, but together they have given me an informal education in what makes design work and what makes it fail.

My partner jokes that there is only one thing I reliably call beautiful: a good streetscape. Whenever we travel, I find myself looking down at kerbs, footpaths, drainage and materials, trying to understand why a place feels right, or why it doesn’t.

Somewhere between infrastructure projects and software teams, I realised that building software for the built world is the work I want to spend my life doing. Northing is where those two halves of my career come together.

Its mission is simple:

Help built-world teams design together and build better places.

A line in a model is not the thing in the ground.

I started working on infrastructure projects at 18. On construct-only contracts, designs usually arrived as PDFs and lightweight models. Working with specialist surveyors, my job was to test them against the site and help make them real.

On one project, an existing water main ran directly beneath a proposed pile. The service had been picked up and was in the model, but only as a single line. Its 600-millimetre diameter and required clearance were not visible. I missed the clash until the piling crew was ready to begin.

We caught it before drilling, but moving the pile meant revising the surrounding design and waiting for approval while crew and plant costing roughly A$15,000 a day stood by.

The lesson was not that someone should have looked harder. Having the information is not the same as being able to understand it.

After that, I spent more time sitting with surveyors, turning existing services into three-dimensional models and checking them against the proposed design. It helped, but it was slow, specialist work that was difficult to repeat under normal project pressure.

When project information conflicted with the site or could not be built as drawn, we raised requests for information, or RFIs. On projects I worked on, they could run into the hundreds. Each pulled the delivery team, designer and client back into a decision after change had become expensive.

Design is a team process, not a handover.

Projects often moved like a relay: survey to design, design to review, review to tender and tender to delivery. Files travelled more easily than the reasoning behind them, so delivery knowledge often arrived after design decisions had hardened.

Involving delivery teams earlier is progress. But earlier involvement only works when the people delivering a project can understand its design and contribute what they know while change is still possible.

Building software showed me a different design process. Product managers, designers and engineers have different responsibilities, but they keep the work visible, explore alternatives and revise it together as their understanding grows. The work moves back and forth, rather than down a chain.

In Let My People Go Surfing, Yvon Chouinard writes, citing Dr Michael Kami:

“Only about 10% of a product’s costs are incurred during the design phase, but 90% of the costs are irrevocably committed.”

Yvon Chouinard, citing Dr Michael Kami

He was writing about product development, but infrastructure makes the principle physical. When project knowledge comes together late, the cost appears as specialist time, rework and programme delay.

Working in infrastructure and software taught me the same thing: design is not a drawing or model produced by one specialist. It is the shared process through which a team decides what should be built and how its parts should work together.

The model should carry the whole design.

The water main was in the model, but its meaning was not.

As I traced that problem upstream, I found a broader pattern. Important civil geometry may be modelled in three dimensions, while drawings, BIM data and delivery outputs are produced through separate workflows. They describe the same project, but meaning and changes do not always travel between them.

Drawings are necessary. They should be views of the design, not separate versions of it.

Northing’s bet is that a kerb, footpath or pipe should remain the same thing as it moves through the project. It should carry through plan, profile, section, 3D, drawings, BIM and delivery outputs.

Designers should be able to make a change once, have it carry into every output, and give the whole project team the same underlying design to understand and improve.

A project should come together in a model before it comes together on site.

Northing will start where infrastructure design is created. Designers should be able to model roads, kerbs, footpaths, services and other civil elements once, then produce drawings, quantities, BIM data and delivery outputs from the same design.

That model should then become a shared project space. Designers author it, clients review it, delivery teams work from it, and surveyors add what they learn. Everyone can understand the same project without reconstructing it from separate files.

Northing will work towards this in three phases:

1. Phase 01
   Model infrastructure projects once.

   Help designers create meaningful civil elements and edit them from the view that best suits the decision.

2. Phase 02
   Let the whole project team work from the same model.

   Connect designers, clients, surveyors and delivery teams so they can understand, question and improve the project while change remains possible.

3. Phase 03
   Carry approved intent into autonomous delivery.

   Make trusted, human-approved design information usable by machine-control systems, site robots and other automated tools.

Machines can execute approved instructions, but they should not decide what gets built or why. As infrastructure becomes more automated, the model will need to carry geometry, meaning, sources, constraints and human approval. Judgement, approval and responsibility must remain with people.

Most people will never see the model behind a project. They will experience its decisions through the footpaths, streets and places around them, often for decades.

When those decisions come together, a place feels coherent. Everything belongs.

I’m building Northing to help more teams create places like that.

Will Smith
3 August 2026