Emoji Jev
September 17, 2026 ยท View on GitHub
Emoji autocomplete at the speed of typing. Live at emoji-jev.whop.site.
Type anything and a keyboard of up to 254 emojis lights up as you go. Every pause in typing is one call to Jev, which answers eight typed questions in parallel in about 100 ms of model time:
- a Choice over 64, 128, or 254 emojis (switchable on the page), with a probability for every key
- a Choice over eight primary emotions, shown as a distribution
- three Scores: mood, urgency, and energy
- three Booleans: sarcasm, joke, and whether the writer wants a reply
Jev bills input tokens only, so a keystroke costs about $0.0002.
Stack
- TanStack Start on Cloudflare Workers, hosted by Whop (
*.whop.site) - AI SDK
experimental_evaluatewith the@ai-sdk/typesafe-aiprovider - No UI framework, no Tailwind: one small stylesheet and the system font
Run it yourself
npm install -g @whop/cli # or: curl -fsSL https://whop.com/install.sh | sh
whop login
whop apps init --template app_D0f07WnZVO8v5N --app_type website # clone the published source
cd emoji-jev
# one of the two, whichever you have access to:
whop apps secrets set --secret AI_GATEWAY_API_KEY=... # Vercel dashboard -> AI Gateway -> API keys
whop apps secrets set --secret TYPESAFE_AI_API_KEY=... # https://console.typesafe.ai/settings/keys
echo "AI_GATEWAY_API_KEY=..." > .dev.vars # same key for local dev (gitignored)
whop apps dev
whop apps deploy
Or from this repo: pnpm install, add .dev.vars, pnpm dev.
Files
src/emoji.tsthe emoji vocabulary and the three questionssrc/jev.tsthe Jev call; uses TypeSafe directly ifTYPESAFE_AI_API_KEYis set, else Vercel AI Gateway viaAI_GATEWAY_API_KEYsrc/routes/api/react.tsPOST /api/reactsrc/routes/index.tsxthe page