Dark Web Crawler & ML Classifier
Machine Learning
Web Scraping
Data Engineering
A sophisticated Flask-based dark web crawler that discovers and classifies .onion sites using machine learning.
Features multi-process crawling via Tor, real-time content classification into 7 categories (93.5% accuracy),
and an interactive graph visualization of 84,000+ indexed sites.
Key Features:
- Multi-process crawling architecture with Tor SOCKS5 proxy integration
- Real-time ML classification (Logistic Regression + TF-IDF) without storing HTML
- Interactive force-directed graph visualization (vis.js) of dark web network topology
- RESTful API with scheduled crawling and job management
- MySQL database storing 84K+ nodes and 111K+ edges with relationship tracking
Technologies & Languages:
Python
Flask
Scikit-learn
NetworkX
MySQL
Tor
Docker
APScheduler
vis.js
Homelab Infrastructure & Automation
Infrastructure
DevOps
Automation
A comprehensive homelab infrastructure featuring automated deployment and configuration management for
self-hosted services. Built using Infrastructure as Code (IaC) principles with Terraform for provisioning
and Ansible for configuration management on Proxmox VE. Includes CI/CD pipelines with self-hosted GitHub Actions runners.
Key Features:
- Automated VM provisioning using Terraform on Proxmox hypervisor (9 VMs)
- Configuration management and orchestration with Ansible playbooks
- Containerized application deployment with Docker and Docker Compose
- CI/CD with self-hosted GitHub Actions runners on each VM
- Nginx reverse proxy with SSL/TLS (Let's Encrypt + mkcert for local services)
Technologies & Languages:
Terraform
Ansible
Proxmox VE
Docker
Nginx
GitHub Actions
Python
YAML
HCL
Linux/Ubuntu
Self-Hosted Reverse Proxy with Dynamic DNS
Networking
Security
DevOps
Production-grade Nginx reverse proxy serving multiple applications with automated SSL certificate management
and dynamic DNS updates. Handles both public-facing domains and local network services with proper SSL/TLS
termination and security headers.
Key Features:
- Automatic SSL/TLS certificate management with Let's Encrypt and Certbot
- Local HTTPS using mkcert for trusted self-signed certificates
- Dynamic DNS integration with ddclient for automatic IP updates
- Proxying 7+ services with custom domain routing and health checks
- Security headers, rate limiting, and HTTP/2 support
Technologies & Languages:
Nginx
Let's Encrypt
SSL/TLS
Docker
ddclient
mkcert
DNS