User repository on GitHub
Can a model design a dashboard? A console whose whole design system is generated at runtime by Jev from a one-sentence brief.
- Forks
- 4
- Last pushed
- Updated Sep 2026
Repo stats from the GitHub API, cached Oct 2026.Homepage
Jev directory
CommunityJev Demos & Experiments
Jev Design
A Next.js console whose whole design system is generated at runtime by Jev from a one-sentence brief: type 'an 18th century printed book' and every visual decision is chosen, not written, from a catalog a person authored.
- Stars
- 59
- Language
- TypeScript
- Reading time
- 3 min
- Published by
- Community
- Author
- Gonzalo Bilune
- Last verified
- Oct 2026
Works forJev for Classification
Used inJev for Software & DevTools
- Languages
- TypeScript
- Platforms
- Next.jsVercel AI Gateway
- Capabilities
- Open sourceRequires API key
Highlights
- Type 'an 18th century printed book' and the console becomes one: aged paper, zero-chroma ink, small-caps labels, hatched fills, square corners.
- Runs on Next.js, Tailwind v4, and shadcn/ui, with the design engine in src/design/ governing tokens, families, and a runtime consistency audit.
- Two routes reach the model: the TypeSafe SDK by default, or JEV_ROUTE=gateway via Vercel's AI Gateway; the adapter renames three fields.
- The recorded demo in docs/demo.mp4 runs six briefs in 83 seconds, in real time, pauses included.
- Vercel WAF rules meter POST /api/design/generate: 3 generations per visitor per 10 minutes and 60 across everyone, about USD 0.25 an hour at the ceiling.
Quickstart
bash
npm install
cp .env.example .env.local # then fill in TYPESAFE_API_KEY
npm run devWatch out
Copy reviewed against the repo README on 2026-10-15; the manifest snapshot lists no license, benchmarks, or pricing. The console runs without a key but POST /api/design/generate answers 503 until TYPESAFE_API_KEY is set, and the README notes the WAF rules are configuration a clone does not inherit.
Questions about this entry
Short answers drawn from the linked project or page.
An experiment asking whether a decision model can design a dashboard. It is a logistics operations console built with Next.js, Tailwind v4, and shadcn/ui whose entire design system is chosen at runtime by Jev from a one-sentence brief.
Jev answers typed questions instead of writing text, so it cannot invent a value: it picks from a catalog a person authored. Every visual and structural decision is declared in the design engine and applied from the outside, so components keep their behavior.
Run npm install, copy .env.example to .env.local, fill in TYPESAFE_API_KEY, then npm run dev. The key stays on the server, and POST /api/design/generate answers 503 until it is set; the console itself runs without one.
By default the TypeSafe SDK calls api.typesafe.ai with your key. Setting JEV_ROUTE=gateway sends the same questions through Vercel's AI Gateway on AI_GATEWAY_API_KEY instead, moving billing and observability to the Vercel account.
Two Vercel WAF rules on POST /api/design/generate: three generations per visitor per ten minutes and sixty across everyone in the same window, a cost cap worth about USD 0.25 an hour at the ceiling. Counters are regional, and a clone does not inherit the rules.
In src/design/: tokens.ts holds the values, families.css declares what shares what, system.css carries flags and per-component adjustments, and design-provider.tsx writes tokens onto the html element as custom properties and data attributes.
Gonzalo Bilune, listed via madewithjev. The README asks whether the style is recognisably the brief and whether it survives the whole component catalog; whether the dashboard looks good is a different, easier question it does not score.
More Jev resources like this
Vercel Labs example that routes three forms by context with Jev and AI SDK: Jev's team choice is accepted when its confidence is at least 95%, otherwise openai/gpt-6-luna-fast independently evaluates the submission and its choice becomes final.
Jev Demos & Experiments#community#catalog-import#logicrw
A Next.js and MUI showcase of Jev classifying Enron-Spam emails: one call asks a Noul spam question, a Choice category question and a Score urgency question at once. Includes a dataset view, a no-key example tab and a stand-alone Python script.
Jev Demos & Experiments#community#catalog-import#logicrw
Community
harsha89/jev-demoTypeScript
harsha89/jev-demoTypeScriptJevSpeak explores whether a model that cannot generate text can still talk: about 13 parallel Jev decisions per turn become a typed Semantic IR that a deterministic compiler renders in English or Chinese. Live at jevspeak.org.
Jev Demos & Experiments#community#catalog-import#logicrw
Community
mm-sheng/jevspeakTypeScript
mm-sheng/jevspeakTypeScriptA solo mahjong coach that takes a 14-tile hand and has Jev recommend which tile to discard, with efficiency, safety and wait scores plus a Japanese explanation, beginner-to-advanced lesson tracks, and a practice mode against three CPUs. Progress and consultations live in Turso.
Jev Demos & Experiments#community#catalog-import#madewithjev
Community
smilior/kiru-hai-coachTypeScript
smilior/kiru-hai-coachTypeScriptAn interactive Next.js workbench for Jev's three decision primitives through OpenRouter: rank 95+ categories with one Choice pass, test binary Noul decisions with transfer curves, and grade inputs on ordinal Score rubrics.
Jev Demos & Experiments#community#catalog-import#logicrw
A BYOK workbench for practicing Jev concepts without official access: emulates State, Noul, Score and Choice via Groq, OpenAI, Anthropic, Gemini or local Ollama, or runs against native Jev with authorized credentials.
Jev Demos & Experiments#community#catalog-import#logicrw
From the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.