Early access · Austin, 2023

Deploy containers without Kubernetes complexity

Nimbus is the container platform for engineering teams of 10–100. Design services visually, deploy with health checks and startup order, and scale on real demand — without operating a cluster yourself.

  • Kubernetes-compatible
  • AWS, GCP, Azure
  • OCI images

production · us-east-1

healthy

api · ghcr.io/acme/api:1.8.2

p50 42ms

Works with the containers, registries, and cloud accounts your team already runs

  • Docker & OCI
  • AWS
  • Google Cloud
  • Azure
  • GitHub
  • Kubernetes API

The problem

Kubernetes is powerful. Managing it shouldn’t be a full-time job.

  • 01

    YAML complexity

    Manifests, Helm values, probes, and ingress rules pile up until a deploy depends on the one person who knows the cluster.

  • 02

    Operational overhead

    Upgrades, networking, node pressure, and noisy alerts become a second job — usually for whoever last touched the infrastructure.

  • 03

    No platform engineers

    Teams of 10–100 want containers, not a full-time platform hire. Most of them should not need one.

Product

The interface in front of your containers

Nimbus sits between your repo and a Kubernetes cluster in your cloud account. You work with services, environments, and deploys. The cluster stays underneath — compatible, and mostly out of the way.

  • Services, not manifests. A graph of containers and dependencies, with defaults you can still edit.

  • Environments that stay isolated. Dev, staging, and production. Secrets do not leak across them.

  • A deploy you can explain. Health, startup order, and a rollout a new teammate can read.

rollout · web · production

18s

  1. 01Pull imageghcr.io/acme/web:1.8.2Done
  2. 02Health checkGET /healthzDone
  3. 03Wait for postgresdependencyDone
  4. 04Shift trafficrolling · 2 replicasLive

Features

Everything between git push and a healthy service

Opinionated defaults for the work a platform team would otherwise build. Nothing here requires you to write a chart.

  • Visual Service Designer

    Draw services and the edges between them. Nimbus reads the graph and starts postgres before anything that needs it.

    edge
    web → api
    data
    api → postgres
    worker
    worker → postgres
  • One-Click Deploy

    Health checks, startup order, and rollouts are the default. Production is one click, not a checklist someone might forget.

    probes
    passed
    startup
    after postgres
    rollout
    2 / 2
  • Auto-scaling

    Scale on CPU, memory, or requests, and down to zero when idle. Set a floor and a ceiling so a spike cannot run away.

    metric
    requests
    replicas
    0 – 8
    idle
    scale to zero
  • Built-in Observability

    Logs, metrics, and traces come with the service. When a deploy misbehaves, you start there instead of opening another tool.

    p50
    42 ms
    errors
    0.2%
    traces
    per service
  • GitOps Integration

    Deploys are commits. Promote staging by merging, and roll production back by shipping the earlier revision instead of a cluster console.

    production
    a1c4e9
    staging
    88ab12
    rollback
    prior commit
  • Multi-environment + Secrets

    Dev, staging, and production stay isolated. Secrets live with the environment, and dev cannot read production.

    dev
    4 keys
    staging
    7 keys
    production
    isolated

How it works

Three steps. No cluster ticket.

  1. 01

    Design

    Compose your services visually. Name the container, the port, the health check, and what has to be up first.

  2. 02

    Deploy

    Push to production with one click. Nimbus waits on dependencies, checks health, and rolls out without dropping traffic.

  3. 03

    Scale

    Auto-scale based on real demand. CPU, memory, or requests — and scale to zero when nothing is calling.

Architecture

Kubernetes-compatible under the hood

Standard OCI images. Your cloud. A workload you can take with you if you ever leave.

Your repo

Git + OCI image

Nimbus

Design, deploy, scale

Your cloud

Kubernetes on AWS, GCP, or Azure

  • AWS
  • GCP
  • Azure
  • Your cloud account

    Run on AWS, GCP, or Azure. The control plane is ours. The cluster stays in the account you already pay for.

  • Standard OCI images

    If it runs in Docker, it runs on Nimbus. Bring the image you already build in CI. No custom runtime.

  • Portable if you leave

    Workloads stay Kubernetes-shaped. Export the spec. You are not stuck on a proprietary scheduler.

Who it’s for

Built for the team that would otherwise become the platform team

  • Startups moving past Compose

    Docker Compose got you to the first customers. Nimbus is the step before you adopt a cluster nobody has time to run.

  • Teams without a platform group

    Product engineers own services end to end. Nimbus covers the defaults a platform team would otherwise spend a year building.

  • Faster microservices, on purpose

    Split a service when you are ready. Ship it without a week of ingress, probes, and pipeline glue.

Pricing

Simple pricing while you’re early

Start free. Upgrade when production traffic shows up. No seat minimums and no enterprise contract.

  • Developer

    For trying Nimbus on a side service.

    $0/ month

    Free for early projects

    • 2 services
    • 1 environment
    • Shared build minutes
    • Community support
    Start for free
  • Team

    Most teams

    For a product team running real environments.

    $49/ month

    Billed monthly

    • Unlimited services
    • Dev, staging, and production
    • Logs, metrics, and traces
    • Git deploys and secrets
    • Email support
    Request Team access
  • Growth

    When traffic and the service count pick up.

    $199/ month

    Billed monthly

    • Everything in Team
    • Request-based autoscale and scale-to-zero
    • Higher build concurrency
    • Priority support
    Talk to us

Early-access pricing. Usage beyond included build minutes is shown before you owe anything. Cancel before you cut production over.

Get started

Get started with Nimbus Container

Tell us about your team. We open early-access workspaces in small batches from Austin — free to start, including the Developer plan.

  • Bring the OCI images you already build.
  • Dev, staging, and production without a platform hire.
  • Kubernetes underneath, on AWS, GCP, or Azure.
Team size
What you run today

No credit card. We reply from Austin, usually within a few days. Early-access workspaces open in small batches.