GPU compute,
signed unsigned
complexity.

A Kubernetes-native AI/ML platform. Deploy inference endpoints and training jobs on GPU clusters with scale-to-zero, zero-trust networking, and no vendor lock-in — every layer is open source and replaceable.

unsigned deploy — session ● live
$ unsigned deploy --model llama-3.1-70b --gpu a100 --precision bf16 --replicas 0:8
Resolving model registry...
Image pulled from Harbor (cached)
GPU pool: 8x A100-80GB MIG (3g.40gb partitions)
Autoscaler: KEDA + Kueue (0 → 8 replicas)
Network: Cilium eBPF + WireGuard mTLS
Endpoint: llama-70b.unsigned.gg

Deployed. Scale-to-zero active. First request wakes in ~4s.
$
idle cost $0 pod-to-pod WireGuard mTLS rollback git revert precision bf16 / int8 API OpenAI-compatible exit path fork the stack endian little, like nature intended
0x01

Infrastructure that gets out of your way.

/01

Scale to zero, scale to thousands

KEDA + Kueue autoscaling with MIG-isolated GPU partitions. Pay nothing when idle. Burst to full cluster capacity on demand. No cold-start tax beyond the first request. Replica counts are unsigned — there is no going below zero.

/02

Zero-trust by default

Cilium eBPF dataplane with WireGuard-encrypted pod-to-pod traffic. Network policies enforced at the kernel level. mTLS everywhere, no sidecars.

/03

Full observability stack

Prometheus, Grafana, Loki, and Jaeger pre-configured. GPU utilization, inference latency, and cost dashboards out of the box. Alert on what matters.

/04

Secrets & policy as code

Vault for secrets management, OPA Gatekeeper for admission control, External Secrets Operator syncing credentials. Nothing hardcoded, ever.

/05

GitOps everything

ArgoCD manages the full stack. Every change is a pull request. Every deployment is auditable. Rollback is a git revert.

/06

OpenAI-compatible endpoints

NVIDIA Dynamo + Triton inference serving. Drop-in replacement for the OpenAI API — bring your own models or pull from the registry. One endpoint, any framework.

0x02

Production-grade, opinionated stack.

Every component is open source. The platform is the integration, not the lock-in.

Edge & network plane
IngressTraefik
API GatewayKong
NetworkCilium eBPF
EncryptionWireGuard mTLS
Control plane
OrchestrationKubernetes
GitOpsArgoCD
IaCTerraform
AuthKeycloak
SecretsVault + ESO
PolicyOPA Gatekeeper
Inference plane
GPU ServingDynamo + Triton
AutoscalingKEDA + Kueue
RegistryHarbor
Observability plane
MetricsPrometheus + Grafana
LogsLoki
TracingJaeger
0x03

From request to inference.

Every layer is replaceable. No proprietary glue. Fork the stack and run it yourself.

Request Flow Client │ ▼ Traefik (TLS termination, rate limiting) │ ▼ Kong (API gateway, auth, quota enforcement) │ ▼ Cilium (eBPF network policy + WireGuard encryption) │ ├─── Control Plane API (Go, manages deployments/scaling) │ │ │ ▼ │ Keycloak (OIDC, RBAC, tenant isolation) │ └─── Inference Plane │ ▼ KEDA (scale 0 → N from queue depth / HTTP RPS) │ ▼ Kueue (GPU quota, fair scheduling, preemption) │ ▼ Dynamo + Triton (model serving, batching, MIG isolation) │ ▼ A100 / H100 GPU (MIG-partitioned, scale-to-zero)
0x04

Pay for compute, not complexity.

No platform fees. No egress charges. No surprise bills. You pay for GPU-seconds consumed — and scale-to-zero means idle costs exactly $0 — an honest, unsigned zero.

  • Per-second GPU billing
  • Scale-to-zero (no idle cost)
  • Free ingress and egress
  • Full observability included
  • No seat-based pricing
  • Open source stack — exit anytime
Rate card

Per-second rates publish with the public beta. Founding teams lock their rates in before launch — that lock survives GA.

A100 80GB · MIG 3g.40gb + full
H100 80GB · MIG + full
CPU · general purpose

0x05 · Early access

Get on the list.

We're onboarding a small group of teams for the private beta. Founding users get locked-in pricing and direct access to the engineering team.

Request Access

opens your mail client — tell us your team, workload, and GPU needs

$ cat request-template.txt
team & company
workload (inference / training / both)
models & expected QPS
current GPU spend, if any