Paths that exist before you need them.
We build private, encrypted connections between your systems, your AI traffic and your providers — and we build the backup paths at the same time, then keep proving they work. Nothing of yours listens on a public address, and nothing has to be improvised at two in the morning.
Who needs this: companies whose AI traffic includes customer records, clinical or financial data, or anything with a residency clause attached — and companies who have already had one bad afternoon caused by somebody else's outage.
THE PATH, DRAWN
WHAT THE ENGAGEMENT INCLUDES
- private overlay
- Mutually authenticated links between your services, with no public ingress on any managed node.
- relay hops you choose
- You nominate the regions and the path; we report per-link latency against it.
- sub-second failover
- Standby paths are probed continuously and switched automatically when one degrades.
- audit log
- One entry for every route change, automatic or manual, with the cause attached.
The same router table, one layer down.
Before GhostRouter routed model traffic, it routed packets. The vocabulary did not change when the payload did: a route is still a match plus a target plus a fallback, a policy is still a file, and a path you have never used is still a path you should have already tested.
Underneath the model layer, GhostRouter runs a private overlay between your services — mutually authenticated links, no public ingress, relay hops you choose, and failover paths that exist before you need them.
Watch a relay degrade and traffic move.
This panel is live, not a recording. Relays degrade, traffic reroutes mid-flight, the log writes the event — or pull the plug yourself below.
RUN THE DRILL YOURSELF
Press and hold — or focus one and hold Space. Probe misses count to three, then the relay drops and traffic moves. Release early to cancel; the automatic scheduler waits while you are driving.
EVENT LOG
What the overlay guarantees, in the words we would use in a contract.
Every promise names its mechanism and the thing you can check.
| Guarantee | Mechanism | Observable |
|---|---|---|
| mutual auth per link | both ends verify identity per session; relays are not trusted intermediaries | link identity in every audit row |
| no public ingress | services dial out only; no inbound listener is created on any node | zero open ports on managed nodes |
| path failover < 800ms class | standby paths are health-probed continuously and switched on probe failure | event log entry per shift |
| audit log per route change | every policy version bump and every automatic shift is recorded with cause | GET /v1/routes/events |
| no plaintext at relay | end-to-end session keys negotiated between the endpoints, not the relay | relay holds no decryption material |
| region pinning | route-level residency constraint refuses targets outside the named region | 403 residency_violation |
Latency figures elsewhere on this page are class targets under a normal operating envelope, not a service-level guarantee. Contractual commitments live in your agreement, where they belong.
Ask about overlay deployment.
Most deployments start with two services and one relay, then grow by adding route entries rather than infrastructure. Tell us which two systems need to talk and we will tell you what it takes.