What it is
An on-demand crypto market intelligence terminal for traders. It ingests 50-1,000 tweets per search, combines them with live price, volume, RSI, funding rate and open interest, and produces a Buy/Sell/Hold/Take Profit decision card. It does not execute trades.
How it uses Jev
Jev evaluates 4 typed questions concurrently on the combined state: Trade Action (Choice: Strong Buy to Strong Sell), Sentiment Spectrum (Score: Extreme Panic to Euphoria), Squeeze Risk (Noul: probability that negative funding plus panic indicates a short squeeze), and Catalyst Significance (Score: None to Major). Results appear as a decision card with confidence percentage, entry range, stop loss, and targets.
Primitives:choicescorenoul
Technique worth stealing
Two-tier pipeline: Python statistical pre-processing plus early-stopping SQLite deduplication to cut API cost before Jev evaluation.
Try it
uvicorn app.main:app --reload --port 8000 after setting TYPESAFE_API_KEY and TWITTER_API_KEY in .env.
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-centricyes0.77
- Shows a System One patternunclear0.48
- Handles uncertaintyno0.04
- Measuredno0.05
- Runnableyes0.62
- Worth recommendingunclear0.45
- Model replicano0.04
- Problem scopescore on a 0–2 scale0.99
- About Jevyes0.98
Signals by Jev jev-1.13.0, card written by DeepSeek V4.1 Flash from the README on 22 Sept 2026.