AI Commerce Experience

Challenge

Conventional storefronts make people translate an intention ("something for a cold-weather trip") into filters and categories. The concept asks what happens when the interface removes that translation step entirely.

Thinking

Framed as an intent-first commerce surface: one conversational entry point, then a progressively-revealed grid that re-ranks against stated constraints. Merchandising rules stay human-owned; the model only re-orders and explains.

Solution

An intent-led storefront concept with transparent, human-governed AI re-ranking.

Build

Built as a working front-end prototype with a mocked catalogue, an intent parser, a re-ranking layer and an explanation strip under every result so shoppers can see why something surfaced.

Experience

Motion carries the reasoning: cards physically reorder rather than snapping, and the explanation strip animates in beneath the shift so the change reads as a decision, not a page refresh.

Outcome

Designed as a concept to explore whether intent-first browsing can replace filter-first browsing without losing merchandising control. No live deployment, no client, no measured commercial results.

Reflection

Next time I would test the explanation strip earlier — it did more for trust than the re-ranking itself.

Gallery