Independent Product Builder & UX Strategist · AI and ML enterprise tools

I design the moment a person decides whether to trust a machine.

Hover a structureSix things I have built or think in

0of 6 structures surveyed

I came to product design through architecture. The same discipline, applied to a different material, runs through all four of these specialisms.

01/ 04

Trust calibration

Before a screen, there was a site. I learned to design for security and clarity first, so a person could move through a space without second-guessing it. Trust calibration does the same for a model, making its uncertainty safe to act on.

Evidence: ShelfView, HITL Annotations

Selected work

This work is under NDA, so client screens stay private. Each case study below narrates the design process through original visuals drawn from scratch. Full artefacts live behind an NDA gate; request access for the password.

A correction surface for an 85% accurate pipeline NDA

An AI editor that turns uncertain model output into structured training signal.

Human in the loopTrust calibration

Scandit2023 – 24 · Sole IC

Making ML output actionable NDA

Turning raw model output into something an enterprise buyer could evaluate and an operator could act on.

Trust calibrationIA

Scandit2023 – 25 · Sole IC

Designing AI reliability for retail execution NDA

AR, OCR and computer vision for real-time price intelligence, rolled out with Staples Canada.

ARComputer vision

Scandit2022 – 24 · Lead IC

Activation failure was a growth constraint NDA

Making invisible system states visible, checkable and recoverable. 60% less support burden.

OnboardingState modelling

Scandit2023 – 24 · Lead IC

Currently building

Plinthly

A mortgage-readiness tool for first-time buyers. The confidence patterns from enterprise AI, applied to a consumer decision that is mostly fear and uncertainty.

Visit plinthly.homes

The design system

Tokens, confidence colour, type and motion — the system this site is built from.

Open the System page