Continuous Delivery for Kubernetes

Koncrete is cloud-hosted continuous delivery for Kubernetes, powered by ArgoCD.

It is the easiest way to setup GitOps deployments for Kubernetes workflows.

Get started
Backed By

Fastest way to setup Kubernetes CI/CD

Stop wasting time installing, hosting, and maintaining complex CI/CD systems that constantly drain resources.

No Installation
Fast, highly available
GitOps Workflow
Almost any cluster
Works with CI tools

Seamless Integrations

Koncrete integrates with the tools you already love. Github webhooks and Slack notifications come built in.

Get Started

Frequently Asked Questions

What is GitOps

GitOps is the practice of using a Git repository containing infrastructure as code as the single source of truth for all code delivery processes. All changes to deployments go through Git. GitOps workflows bring organization and structure to deployments, yielding faster development cycles. It is also easier to audit changes, roll back, and quickly duplicate environments. Adopting a Gitops workflow unlocks the power of CD tools like Koncrete.


What is ArgoCD

ArgoCD is an open source solution for GitOps based continuous deployment. It was initially written for by the Intuit team for internal use, and is one of the most popular continuous delivery tools for Kubernetes clusters. ArgoCD itself runs on Kubernetes, and connects directly with Git repositories and clusters to manage continuous delivery.


Why use Koncrete instead of self-hosting Argo?

Self-hosting ArgoCD takes time to setup and maintain, and requires cluster resources. Features like slack notifications and image updater are not included out of the box. If you prefer to keep tooling separate from your deployment cluster, you need to maintain a separate cluster just for ArgoCD. Koncrete is a cloud hosted alternative that comes with several additional Argo features immediately configured.


Why don't I just deploy in my CI pipeline?

In theory, a CI server allows you to run any job or script, so you can do anything with them. However, it is frequently cumbersome to perform Kubernetes deployment inside a CI pipeline, as all the details and permission of the cluster have to be exposed. CI pipelines that perform tasks within a cluster directly are characterized as overly complex, sluggish, and difficult to debug.
Having a dedicated operator that watches a Repo to sync a cluster is cleaner and more robust.