All posts
Security, infrastructure, networking and tooling — 185 write-ups.

Why Entropy Failures Haunt Self-Hosted Cryptography
A practical analysis of entropy exhaustion, kernel RNG validation on Linux nodes, and defensive key generation patterns.

Self-Hosted DNS-over-HTTPS: A Practical Setup Guide
Set up DNS-over-HTTPS on your own server to encrypt DNS queries and prevent ISP snooping. A five-step guide covering configuration, testing, and harde…

Running AI Coding Agents on a Remote Hetzner Box via Tailscale
The friction of running AI coding agents on remote hardware is real. pocketdev collapses that gap with a single Tailscale-based deploy on Hetzner, let…

Why AI Agents Will Shrink Their Context Windows
AI agents are drowning in their own context. The future belongs to compression, not expansion — and it changes how we build self-hosted AI infrastruct…

Telegram Proxy Management Done Right in Rust
How a Rust-based Telegram proxy manager solved the config chaos I was dealing with across multiple servers and networks.

Claude Watch: VPN-Aware Monitoring for AI Coding Agents on macOS
How a macOS monitor watches your Claude usage and enforces VPN policies automatically, keeping your AI coding sessions private and controlled.

The Quiet Rise of Terminal User Interfaces
Terminal user interfaces have evolved from boring gray slabs to the preferred interface for infrastructure work. Rust, AI agents, and self-hosting dro…

Mobile Security Hardening: A Red Team Checklist
A practical mobile security checklist for red team operators. Covers device encryption, network traffic inspection, certificate pinning bypass, SIM sw…

Why a VPN Doesn’t Make You Anonymous
A VPN encrypts your traffic and hides your IP, but it doesn't make you anonymous. DNS leaks, browser fingerprinting, and traffic metadata can still ex…

Self-Hosted AI Agents Don’t Automatically Mean Private
Running an AI agent on your own server doesn't guarantee privacy. Data still leaks through telemetry, model APIs, and logging. Here's what actually ha…