Because running containers in the cloud shouldn't be hard.

Per-second compute for stateless services, AI agents and jobs with built-in horizontal scaling.

push an image,
pick a region,
you're live

Terminal

$ cfo cluster create production --regions us-east,eu-west
✓ Cluster 'production' created in 2 regions
$ cfo service deploy api --image ghcr.io/org/api:1.2.3
✓ Service 'api' deployed with 3 replicas
$ cfo service publish api --type load-balanced
✓ Service published at https://api-production.clusterforge.app

Trusted by developers at

"We shipped in an afternoon—no VPCs, subnets, or ALBs." — CDN77 Engineering

Simple primitives, powerful

automations

Everything you need to run
stateless services and jobs at scale.

Simple primitives

Account → Clusters → Services/Jobs/CronJobs. Clean abstractions that map to how you think about infrastructure.

HTTP ingress built-in

Private by default; publish via global LB with path-prefix rules. No ALBs, no VPCs to configure.

Guardrails that stick

Per-cluster quotas so dev can't starve prod. Set hard limits on containers, CPUs, memory, and storage.

Jobs & Cron, first-class

Standard cron semantics with retries, deadlines, and run history. Scheduled work that just works.

Security options

Optional VM isolation, deny-by-default networking, tenant DNS, and encrypted control plane volumes.

Usage-based pricing

Pay for what you use, when you use it. Per-second billing with no upfront costs or commitments.

Simple pricing,
that's always predictable No surprises!

Per-second

CPU & RAM metering

1-minute minimum

Storage

Ephemeral GiB-hours

Per-second billing

Clusters & Services

No object tax

$100

Free credit

On sign-up

How it works

Three steps from code to production. No YAML manifests, no cluster management.

Step 1

Create a Cluster

Pick regions and set quotas. Define your blast radius and resource limits upfront.

clusterforge cluster create production --regions us-east,eu-west --cpu-quota 100 --memory-quota 400GB

Step 2

Deploy a Service

Push your image, set replicas, CPU/memory, and health checks. Zero-downtime deploys included.

cfo service deploy api --image ghcr.io/org/api:1.2.3 --replicas 3 --cpu 1 --memory 1GB

Step 3

Publish & Schedule

Add HTTP ingress or DNS publishing, then schedule CronJobs for background work.

cfo service publish api --type load-balanced --path /api
clusterforge cronjob create backup --schedule '0 2 * * *'

Configuration that makes sense

Declarative YAML that maps to how you think about services. No Kubernetes complexity.

Service Configuration

Complete service definition with health checks and isolation

# clusterforge.service.yaml
name: api
cluster: production
image: ghcr.io/org/api:1.2.3
replicas: 3
cpus: 1
memoryMB: 1024
ports:
  - port: 8080
    protocol: tcp
# Load-balanced for HTTP with health checks
type: load-balanced
healthCheck:
  protocol: http
  port: 8080
  path: "/healthz"
# Stronger sandboxing, optional
isolation: vm

Deploy with: cfo service apply -f clusterforge.service.yaml

Security & reliability

Enterprise-grade security built in. Compliance-ready from day one.

VM Isolation

Optional VM isolation (Kata/QEMU)

Network Security

Deny-by-default network policies

Tenant DNS

Tenant-scoped DNS resolution

Encrypted Transit

WireGuard pod-to-pod encapsulation

Encrypted Storage

Encrypted platform volumes (LUKS2)

Graceful Preemption

Graceful pre-emption with replacement replicas

Built with compliance in mind: SOC 2, GDPR, and enterprise security standards

SOC 2 Type II GDPR Compliant Enterprise Ready