AI · Fraud Detection · Founded 2024

ZeroPredict

AI fraud detection for Web3

ReleasedAI · Fraud DetectionZero-Blockchain venture

ZeroPredict is an AI-powered fraud detection and predictive analytics platform built for Web3. It runs pattern recognition, behavioural analytics, threat monitoring, attack-vector analysis, smart-contract auditing, transaction monitoring and wallet screening - feeding KYC/AML and identity verification surfaces and an inter-app threat sharing network.

1.2M

Events / sec

< 1.5s

Mean detection

< 0.4%

False-positive rate

Models in prod

What it does

Pattern recognition & risk scoring

Anomaly detection at line-rate

Smart contract security auditing

Wallet screening + auto-block rules

KYC/AML + identity verification

Cross-app threat sharing network

Engineering stack

Inside the product

A walkthrough of the most-used modules. Each is a fully-built screen, not a mockup.

Tip: click the active slide to open it full‑screen. Tall screenshots scroll inside the slide.

Dashboard
Fraud Detection
Pattern Recognition
Risk Scoring
Anomaly Detection
Behavioural Analytics
Threat Monitor
Predictive Analysis

Every screen, every flow

12 core modules - tap a tab to inspect what your customers see.

Dashboard

Operational fraud dashboard.

Sector

AI · Fraud Detection

Screenshot gallery

Click any tile to inspect it full‑screen.

How it flows

End-to-end architecture and user flow for ZeroPredict.

Stream

Ingest on-chain & off-chain events at line-rate.

Score

Models score wallets, txns and contracts continuously.

Decide

Auto-block rules quarantine high-risk activity.

Cross-app threat intelligence shared across the network.

Changelog from my journey

I've been working on Aceternity for the past 2 years. Here's a timeline of my journey.

Step 1

Stream

Ingest on-chain & off-chain events at line-rate.

Step 2

Score

Models score wallets, txns and contracts continuously.

Step 3