Jev Design Test
The project turns a single probabilistic request into a varied exploration grid by combining independent slot distributions, enforcing incompatibility rules, and selecting diverse combinations.
The project turns a single probabilistic request into a varied exploration grid by combining independent slot distributions, enforcing incompatibility rules, and selecting diverse combinations.
Given a free-text page request, Jev routes it to a predefined page type and independently returns a Choice distribution for every layout slot across the vocabulary. Application code samples combinations from the selected page's per-slot distributions and renders screens from existing components; Jev does not write JSX or copy.
Which app navigation structure best fits this request?
The request state is {query: <free-text page request>}. Source sends a page Choice over all 18 PAGES ids and routing descriptions, plus per-page/per-slot Choice questions. This example displays the page vocabulary and the dashboard nav slot; it does not represent a live response.
Page types declare a routing description and named layout options per slot; option keys also identify renderer components.
The dev middleware sends the free-text query as state with one Choice question for page type and one for each page/slot pair.
The client builds combinations from the returned distributions, filters excluded slot combinations, and greedily selects a varied set.
Existing UI components render the sampled schemas as page frames on the canvas.
The project turns a single probabilistic request into a varied exploration grid by combining independent slot distributions, enforcing incompatibility rules, and selecting diverse combinations. It keeps generated structure inside a fixed, real component vocabulary.