Autonomous
Digital Forensics & Incident Response

Accelerate threat hunting using the AMD Developer Cloud. Parse raw memory dumps into actionable intelligence instantly with our decoupled Unicorn Architecture.

inzex-engine-v3.exe — [ACTIVE]
Scroll to Explore

AMD Developer Cloud

Powered by high-performance ROCm 7.2 and vLLM 0.16.0 instances for incredibly rapid AI inference without local bottlenecks.

Gemma 4 Intelligence

Automated IOC extraction and structured JSON threat reporting using state-of-the-art local LLMs on AMD hardware.

Volatility 3 Integration

Native Python execution uncovers deeply hidden malware by seamlessly parsing raw memory dump structures.

The Decoupled Unicorn Architecture

Bridging large memory dump handling, AMD accelerated computational forensics, and real-time human review.

Client Tier (Next.js)

UI: Server-Side Rendered Dashboard

Features: File Upload API, Human Review Workspace, AI Chat

1. Uploads .vmem &
Submits comments
6. Live updates:
Threat Reports &
AI Replies

State & Storage Tier (Supabase)

Storage Bucket: cridex.vmem

DB Tables: Cases, Findings, Finding_Threads

2. AMD Worker polls
for new files or
comments
5. Writes JSON
analysis & updated
context

The Unicorn Engine

AMD DEVELOPER CLOUD (ROCm NOTEBOOK)

Python FastAPI app runs Volatility 3 (CPU) & Google Gemma 4 (GPU)