Jev 101
Start Guides
Search
中文 EN
Platform heat 1 · Quiet★ 0heat incomplete (no X)

Jev Lab (Nijal AI)

A MIT FastAPI workshop that lets participants probe hosted TypeSafe Jev against a chat LLM on phone or PC—missions, duels, calibration tables—without writing code.

Repo: Nijal-AI/jev-lab ↗

What it is

Feel Jev vs an LLM in a 45-minute facilitated lab.

What it does: Hosts scenario probes, threshold missions, and head-to-head duels with latency/cost/calibration panels.

Why it belongs in the showcase: Teaching surface for real hosted Jev decisions with optional live keys.

Caveats: Facilitator tool; sim mode without a key.

Evidence

Public MIT repo, created 2026-10-28, 0 stars as of 2026-10-29. README (French) documents TYPESAFE_API_KEY (empty → simulated mode), JEV_MODEL default jev-latest, Discover/Missions/Duel/Admin tabs, anti-scam filter with probability thresholds, ticket routing, and calibration vs an OpenAI-compatible LLM. HTTP 200 on GitHub.

Caveats

  • Without TYPESAFE_API_KEY the lab runs in simulated mode
  • UI copy is primarily French; not an official TypeSafe product
  • Per-participant call quotas and admin code are operator-configured

Unofficial case page. Stars / list position observed for heat; do not treat heat as a quality SLA. Review the upstream README before you depend on it.