Networks & Freedom

I grew up behind filters. Where I live, parts of the internet are routinely unreachable — not because servers are down, but because someone decided they should be. That reality turned networking from a hobby into a survival skill, and the survival skill into a specialty: building infrastructure that keeps people connected.

My work in this area is practical rather than theoretical. I design and operate VLESS tunnels and proxy panels, manage multi-user subscription systems with Clash, v2ray and sing-box compatibility, and deploy edge infrastructure across Cloudflare Workers, Hugging Face Spaces, Railway, Render, Fly.io and Koyeb. Each of my panel projects — Usf Panel, usf-edge, edge-pp, Render-usf — attacks the same problem from a different platform, because blocking is an arms race and single points of failure get people cut off.

What I actually build

• Tunnel panels — multi-user VLESS management with subscription links, traffic stats and one-command deploys.
• Edge deployments — proxy state on Cloudflare KV across 300+ PoPs, zero cold starts, no servers to seize.
• Diagnostics — pure-Go tooling like NetPilot for port scanning, TCP ping and HTTP health checks.
• Resilience patterns — multi-platform redundancy so one takedown never equals zero access.

Why this matters

In an environment where connectivity itself is contested, a proxy panel is not a gadget — it is a lifeline for work, education and free information. I treat it with the seriousness of critical infrastructure.

The mindset

Networking taught me how systems fail: gracefully, partially, and always at the worst moment. It taught me to read packet captures the way other people read error messages, to distrust "temporarily blocked", and to always have a second path, a third fallback and a fourth idea. Those instincts bleed into everything else I build — redundant APIs, failover model routing, and panels that assume the network is hostile.

If your project needs to survive a hostile network — or you just want your stack to stop falling over when a region goes dark — that is exactly the kind of problem I enjoy most.