What it is
Jevtown lets people post a text, listing, product or headline and watch a simulated town of 10,000 AI personas react within seconds — most scroll past, some like, repost, block, write to the seller or buy.
How it uses Jev
Jev answers typed questions with probabilities and writes no text. One request asks 60 Score questions on who would care plus seven Noul moderation checks (0.5 keeps text out of the public feed, 0.85 blocks it). Batched Choice questions then return each persona's reaction in waves of 600, 1,500 and 3,000; code sends the text to the next wave only while glad reactions outweigh sorry ones by at least 0.1 of the wave.
Primitives:choicescorenoul
Technique worth stealing
Wave-based audience simulation: propagate a text through persona waves only while glad reactions outweigh sorry ones by a fixed margin.
Try it
npm install, cp .env.example .env.local, paste a Jev key, npm run dev (http://localhost:5191).
Evidence
Each line is one question put to Jev about the README. ≥ 0.60 reads as yes, ≤ 0.40 as no; in between Jev is not making a call.
- Jev-centricyes0.96
- Shows a System One patternyes0.96
- Handles uncertaintyunclear0.40
- Measuredno0.15
- Runnableno0.34
- Worth recommendingunclear0.58
- Model replicano0.05
- Problem scopescore on a 0–2 scale0.47
- About Jevyes0.97
Signals by Jev jev-1.13.0, card written by DeepSeek V4.1 Flash from the README on 22 Sept 2026.