# Fraud detection with Jev and Kimi K3

Jev sorts 100 emails in 1.42 s and sends the unsure ones to Kimi K3.

Grade: A — A clean typed decision \(yes/no, a pick from a list, or a level on a scale\) that is concrete, repeats routinely, and has meaningful value.. Score: 4.12.

Vertical: Finance & insurance.

Function: Agents & dev.

Status: Seen in the wild.

Volume: routine.

Value: meaningful.

Risk: high.

Evidence: built and shown.

Flags: human-review.

- [Source](https://madewithjev.com/builds/fraud-detection-jev-kimi)
- [Vertical](/verticals/finance_insurance/)
- [Function](/functions/agents_dev/)
- [Jev System One AI model for instant decisions](/use-cases/jev-system-one-ai-model-for-instant-decisions/)
- [IP address geolocation fraud](/use-cases/ip-address-geolocation-fraud/)
- [Weekly account pass: is an anomaly worth a look](/use-cases/weekly-account-pass-is-an-anomaly-worth-a-look/)
- [40,000 wallets screened in four seconds](/use-cases/40-000-wallets-screened-in-four-seconds/)
- [Fraud Detection with AI: Ensemble of AI Models Improve Precision & Speed](/use-cases/fraud-detection-with-ai-ensemble-of-ai-models-improve-precision-speed/)
- [jev-trader](/use-cases/jev-trader/)