Malware detection using LLM

Built a PoC system to detect malware using sandboxed analysis, fine-tuned LLMs, and RAG.

Python
Django
Docker
AI Security

Project Overview

In this project, I built a Proof of Concept (PoC) system designed to detect malware effectively. The system leverages sandboxed analysis to safely execute and observe potential threats. By integrating fine-tuned Large Language Models (LLMs) and Retrieval-Augmented Generation (RAG), the system improves detection accuracy and provides detailed threat analysis.

Technologies Used: Python, Django, Docker.

Key Learnings: AI-based threat detection, secure model integration, and data handling.

Detection Interface

The web interface provides real-time insights into the analysis process, displaying classification results and threat intelligence extracted by the LLM.