Sprintsail is a managed application platform. Point it at a directory of source: it works out the language, installs the dependencies, builds it, starts it, and keeps it running. Attach Postgres, Redis, object storage, search, or a queue and the credentials land in the app’s environment. There is no cluster to provision, no YAML to maintain, and no build pipeline to assemble first.
Five steps from an empty account to a running application with a database attached. No cluster to provision, no pipeline to assemble first.
Single sign-on through the platform front door. Your identity carries your roles with it — every action you take is scoped to what you are actually allowed to do.
SSO · no shared admin accountSelf-service onboarding creates your workspace, then separate environments inside it — dev, staging, prod. Create one in either place and it exists in both, with nothing to reconcile.
workspace → environment → appUpload your code from the console or push it from your terminal. The platform works out the language, installs the dependencies, and produces a runnable image. No Dockerfile required.
source inPick Postgres, Redis, object storage, search, or a queue. Binding injects the credentials into your app environment as variables.
bind → env varsLive logs, per-instance metrics, deployment history, and scaling — in the console or your terminal, reading the same platform either way.
logs · metrics · scaleA running app is rarely just a process. It needs a database, somewhere to put files, a queue for the slow work, and a way to see what it is doing at three in the morning. All of it is here, and all of it is self-service.
Push source, get a running app. The platform detects the language, builds an image, and runs it — restarting it when it falls over. Deployment history is kept for every push.
Managed Postgres, provisioned on request and bound to your app with the credentials injected as environment variables. No ticket, no wait.
Managed Redis for cache and session state. The same self-service provision-and-bind flow as every other service here.
S3-compatible buckets provisioned per environment. Bind to an app and the credentials arrive in the environment — your existing S3 client works unchanged.
Managed search and log analytics. Provisioned and torn down through the same self-service path as every other service, so the lifecycle is consistent.
Managed message queues for async work. Provision, bind, and your worker starts consuming — no broker to run yourself.
Real per-instance CPU and memory, read live from the running instances — not a cached summary. You see each instance separately, so a single sick instance is visible.
Live application logs streamed from the running instances, in the console or in your terminal. Same logs either way.
Change instance count and memory and the platform rolls it out. One writer owns the instance count, so the console and the terminal never fight each other.
There is no cluster to size, no nodes to patch, no build system to keep working, and no glue to write between a database and the app that needs it. Sprintsail is managed Kubernetes, and the point is that you never open it: capacity, node health, certificates, routing, service provisioning, upgrades and patching all sit below the line you work at, and none of it is yours to run.
You get a console, a terminal, and somewhere to push. Your environment is dedicated to you, not a shared pool you queue behind. If you later need the same platform running inside your own boundary rather than ours, that is Drydock — the same Kubernetes platform shipped to you instead of operated by us. A different address, not a rewrite.
Deploys, logs, environment variables, memory changes, and service bindings all work from the command line, against the same platform the console talks to — so the two never disagree about what is running. Pick whichever door suits the moment. Compatibility note: the client is the standard cf CLI, so a team that already has it installed has nothing new to install. Full reference in the docs.
Log in once with single sign-on and your workspace and environment targets are set — the same identity and the same roles you have in the console.
# point the client at Sprintsail$ cf api https://api.sprintsail.com$ cf login --ssoworkspace acmeenvironment prod
Push a directory. The platform works out what the app is, installs its dependencies, and produces an image it then runs. No Dockerfile to write and no YAML to maintain — though you can bring your own image if you prefer.
# no Dockerfile required$ cf push orders-apidetected Node.jsbuilding ████████████ donestarting 1/1 instances✓ app started
Stream logs from the running app, or pull recent history. The console reads the same source, so what you see in a browser and what you see in a terminal never disagree.
# follow the running app$ cf logs orders-api[APP/0] listening on :8080[APP/0] GET /health 200[APP/1] GET /orders 200
Change the instance count or the memory limit and the platform rolls the change out. One writer owns the instance count, so console and terminal changes converge instead of fighting. Scaling is a change you make, not one the platform guesses at.
# scale out and up$ cf scale orders-api -i 3 -m 1Gscaling to 3 instancesrolling ███████░░░░░ 2/3✓ 3/3 running
The platform inspects what you pushed, works out how to build it, installs dependencies, and produces a runnable image. If you would rather build your own image, push that instead — both routes end in the same place.
Push a directory. The platform works out what the app is, installs its dependencies, builds an image, and starts it — nothing to write, nothing to maintain, no pipeline to assemble first. Every push is recorded, so you can see what went out and when.
One pushNothing to provision, size, upgrade, or patch. Capacity, node health, certificates, routing, and platform upgrades are ours. You get somewhere to push and a place to watch it run; the Kubernetes underneath is never something you open.
Managed for youPostgres, Redis, object storage, search, and queues are one request away. Provisioning happens for you, and binding puts the credentials into your app environment as variables — alongside the ones you set yourself. Your existing client library reads them and connects, unchanged.
Self-serviceApplication output streams live from the running instances, in the console or in your terminal, and CPU and memory are reported per instance rather than as a rolled-up average — so one sick instance among three is visible instead of averaged away. When an app needs more memory, change it and the platform rolls the change out.
Live, per instanceSingle sign-on at the front door, and every action after that runs as you. Your roles ride along with each request and are enforced per call, so the console can never do more on your behalf than you could do yourself.
SSO · RBACA workspace for the team, and separate environments inside it for dev, staging, and prod. Create one in the console and it is there in the terminal in the same breath; delete it in either place and it is gone from both. There is no second set of books to drift out of date.
Dev · staging · prodSprintsail is open to a small number of teams while we run it in anger. We would rather have a handful of applications we genuinely support than a signup button and a queue of tickets.
Bring an application you actually care about. We onboard you, you push, and we stay on the other end of the line while you do it. In exchange we want honest feedback about what breaks.
cf CLI against the same APIIf the platform needs to live inside your compliance boundary rather than ours, that is Drydock — the same Kubernetes platform, deployed single-tenant in your environment. Same workflow, different address.
Candidates earn a Verified credential through a graded assessment, a lab, and a 45-minute SME session. Hire from a pool that's already proven.
A different door to the same estate. Declare your app in TypeScript, deploy to your own AWS account, and move it later without rewriting handler code.
Early access is open to a small number of teams. Tell us what you want to run and we will get you onboarded.