Core concepts

Organizations#

An organization is the top-level container in NuPaaS. Every resource — projects, databases, domains, API keys — belongs to exactly one organization. Organizations are multi-tenant: each is fully isolated from others.

Organization owners can invite members and assign roles. Roles control who can deploy, manage billing, configure infrastructure, or administer the organization itself. Every action is scoped to an organization, so members only see what they have access to.

Projects#

A project represents a deployable unit on NuPaaS. Each project is linked to a Git repository. When you push code, NuPaaS detects the change, builds your application, and deploys it.

Projects contain one or more services (see below), environment variables, domain bindings, and a deploy history. You can configure build settings, resource limits, and scaling rules per project.

Environments#

Environments let you run the same project in different stages. NuPaaS supports three environment types:

• Preview — automatically created for pull requests. Each PR gets its own isolated URL for testing changes before merge.
• Staging — a long-lived environment for integration testing and QA before promoting to production.
• Production — the live environment serving real traffic on your custom domain.

Each environment has its own set of environment variables, secrets, and scaling configuration. Promotes and rollbacks move builds between environments without rebuilding.

Services#

A service is a running container within a project. Most projects start with a single web service, but you can add background workers, cron jobs, or additional processes as separate services.

Each service has its own resource allocation (CPU, memory), health check configuration, and scaling rules. Services within the same project share the same network and can communicate over internal DNS.

Servers#

Servers are the underlying infrastructure that runs your workloads. NuPaaS manages a cluster of servers on Hetzner hardware in Germany. You do not need to provision or manage servers directly — the platform handles scheduling, scaling, and health monitoring.

For advanced use cases, you can inspect server status, view resource utilization, and configure server groups for workload isolation or GPU-accelerated tasks.