Selected work

Three things I couldn't stop thinking about, so I built them.

The deeper cut: the screens as they were built, the decisions behind them and what remains unproven. SOURCE IT uses demo data with live search not connected. Relay and OFFLINE are working prototypes; neither has been tested with users.

SOURCE IT

Sourcing investigation tool — The Certainty GapBUILT

Similarity isn’t proof. Two products can look alike without sharing a manufacturing origin — so the interface has to show how confident it is, and why.

The problem

Sourcing claims travel faster than the evidence behind them. Someone sees a product, finds a visually similar listing and treats resemblance as an origin match. The gap between what an image shows and what can actually be supported is where the mistakes happen.

The build

A working prototype where a search can begin with an image, a URL, a Pinterest reference or a written description. Results are returned as leads with a stated confidence level and the supporting evidence attached, rather than a single confident answer.

Prototype walkthroughSOURCE IT — prototype walkthrough. Demo data; live search not connected.

Decisions

Resemblance is presented as a lead, not a conclusion.: The whole failure mode of this category is confident matching. Naming the uncertainty is the product, not a disclaimer on top of it.
Confidence and its supporting evidence are shown together.: A percentage alone is just a different kind of claim. Showing what the number rests on lets the person judge it themselves.
Start from the reference someone already has.: A sourcing question usually begins with something in hand — an image, a link, a saved reference, a description. The prototype accepts that as the starting point.

What remains unvalidated

Live search is not connected — the walkthrough runs on demo data. Sourcing accuracy, commercial launch, customers and supplier partnerships have not been validated.

Working prototype — demo data, pre-launch.

Tools

Lovable: — prototype build and interface
Walkthrough: — demo data; live search not connected

Relay

AI conversation studioBUILT

Transcription is only the raw material. Which moments, ideas and quotes actually matter — and who should make that final editorial decision?

The problem

Meeting tools preserve the words, but the useful story is buried inside them. What matters changes with what someone needs next: a decision log, a study note, a follow-up email. Relay finds the relevant moments in the same messy conversation, then keeps the final editorial decision with the person using it.

The build

A multi-screen prototype that runs a live model call end to end — upload, analysis, intent-driven output selection, a review step, then a library of saved outputs.

Relay prototype — the upload screen, where a conversation recording is dropped in for analysis — PrototypeUpload a conversation recording.

Relay prototype — choosing which structured output to generate from the same recording — PrototypeIntent selects the output: the same recording becomes a summary, action items or notes.

Relay prototype — every detected speaker must approve before anything is shared — PrototypeNothing is shared until every detected speaker has approved it.

Decisions

Intent selects the output, not the file type.: The same recording becomes a different artefact depending on who will act on it. Asking for intent up front is the only question that changes the answer.
Nothing publishes without a human edit.: The review screen cannot be skipped. The model drafts; the person keeps the editorial decision.
Output is stored, not consumed.: A library turns one recording into a compounding set of working assets rather than a result that is read once and lost.

What remains unvalidated

Whether stated intent is a better selector than document type once real users are choosing under time pressure.

Working prototype. User validation not yet run.

Tools

Lovable, Replit: — building and iterating the prototype
ChatGPT, Claude: — product logic, flows, output architecture, edge cases

OFFLINE

Real-world social discoveryBUILT

A product that optimises toward real-world interaction, not time spent in-app.

The problem

A match on a dating app opens a long digital runway — opening lines, days of messaging, scheduling — and often no meeting. Meanwhile people are already out in the same room with no way of knowing who is compatible, who is interested back, or whether it is alright to walk over. Three kinds of uncertainty, none of them solved by more messaging.

The build

A multi-screen prototype covering the two contexts the product runs in: a persistent discovery layer outside venues, and Tonight Mode, activated when the user is physically out. Tonight Mode covers temporary venue check-in, a compatibility count rather than a catalogue, a small curated batch, mutual interest, one active connection with minimal coordination, and a 'We've met' confirmation.

OFFLINE prototype — the Tonight Mode entry screen, turned on when the user arrives somewhere — PrototypeTonight Mode entry — switched on when you arrive, not left running.

OFFLINE prototype — the live venue list showing how many people are checked in at each place — PrototypeLive venue state — who is out, without displaying everyone.

OFFLINE prototype — a checked-in venue showing a compatibility count and a countdown to when the night ends — PrototypeA compatibility count rather than a catalogue, and a connection that ends with the night.

Decisions

One active connection at a time.: Several mutual likes in one venue would otherwise become several parallel chats and no conversation. Mutual interest becomes a sequenced opportunity with a short window, so the night is spent on one person rather than a phone.
A curated batch, never an endless queue.: '37 checked in · 12 compatible' orients without displaying everyone. Showing all twelve at once turns a room into a feed and puts the product in competition with the night.
Approachable Mode is recognition, not permission.: An optional current-outfit selfie answers 'what do you look like right now', which is a different job from the profile photo. It is contextual, temporary and revocable, and it never overrides an ordinary boundary.
Expiry is contextual, not punitive.: An unmet connection ends with the night — 'not tonight' rather than a loss. If two people do meet, both confirm it and the connection is kept.

What remains unvalidated

Whether people want to know which compatible singles are in the room, or whether that reads as intrusive — and whether a deliberately limited batch feels refreshing or restrictive. Both have to be answered before venue density is worth engineering for.

Working prototype. User validation not yet run.

Tools

Lovable, Replit: — building and iterating the prototype
ChatGPT, Claude: — behavioural problem definition, product architecture, flows, trust and safety logic

Back to the story →