01
Own
software
We are developing the first Wharf Machine products now. They are Mac-native, local-first and designed around AI from the beginning. We are not announcing the products yet.
Native software. Local intelligence.
Wharf Machine is an independent software company and production house. We design and build our own products, and lead selected product engagements for other organisations.
Discuss a product
Currently taking a small number of engagements
What we do — 01
Own
02
Product
building
Purpose-built Mac software, focused web applications and agentic systems. We lead the work from product direction and interaction design through prototypes, technical decisions and production software.
03
AI-native
advisory
Helping organisations become genuinely AI-native across product, design, development and day-to-day operations. We identify where AI can increase the pace and ambition of work, then shape usable workflows, evaluation loops and prototypes that teams can adopt.
Local does not mean one device.
It means keeping capability, data and economics close to the product.
Local AI — 02
Local first
Run intelligence on the user’s Mac, a nearby Linux server or small edge hardware whenever the task and hardware allow it.
Open-weight
Build on capable models that can be run, evaluated and replaced without handing the product to one provider.
Cost-aware
Use hosted models only when their additional capability earns the additional operating cost.
Product-native
Put AI inside clear, well-designed workflows—not beside the product as another generic chat box.
Principles — 03
The values behind our own products, commissioned work and advisory.
Design leads
UX and interaction design determine what should be built and how it should feel. They are part of product direction, not a finish applied later.
Native and fast
Use native Mac software when platform fit, speed and interaction quality materially improve the product.
Local and ownable
Prefer systems people can run, understand and keep. Recurring fees should pay for recurring value.