Route optimization API that solves fleet planning as a single problem
Dynamic geography clustering beats fixed zones for large-scale routing problems.
Self-healing WireGuard mesh. Reroutes around failures in seconds, no coordination server, fully FOSS.
Babel routing enables multi-hop WireGuard mesh—Tailscale and ZeroTier only do direct links.
Network engineers, multi-cloud operators, game streamers needing low-latency routing
Tailscale · Nebula · ZeroTier
I also had latency-sensitive "work" ahem (game streaming). So if I were on the same physical network as my gaming pc, I want my VPN to route via the lowest latency LAN path, only falling back to other nodes when I'm travelling or at school.
Note: I have considered Tailscale, Nebula and the like. These work most of the time, but do not give me control over how data is routed. They generally establish direct links (or at most, 1-hop via a relay), and do not take the state of the underlying network into account.
With nylon, I can choose to add links with more premium networks like CN2 GIA or Akamai's (via two Linodes in diff regions). Nylon would take these links into account, and dynamically pick the best routing using Babel (RFC 8966).
If you're interested in the details, I wrote a blog post diving into the challenges of building this: https://jiaqi.dev/posts/nylon
Docs for getting started: https://nylon.jq.ax
Would love to hear thoughts & feedback! Thanks :)
Dynamic geography clustering beats fixed zones for large-scale routing problems.
Managed OSRM API that undercuts Google Maps pricing without the sales calls.
Buzzword-heavy README promises enterprise-grade features but only has 1 GitHub star.
Reverse-engineers free Gemini API; smart quota rotation, but against Google's terms of service.
Single-process routing plus VRP beats VROOM on route quality at equal runtime.
Agent writes missing upload_file() mid-task and commits it — no framework can do this.