Kratix etc.

While an orchestration framework like Kratix aims at unblocking developers through self-service and pre-configured resource setups just like the Humanitec products, it takes a different approach in a number of ways.

The Humanitec Platform Orchestrator is a SaaS solution backed by SLAs, therefore not requiring engineering resources of your own to operate. Kratix needs to be installed in your infrastructure, operated, and hardened, by your engineering team.

The Humanitec Orchestrator comes with an interface choice of API, CLI, or web-based Portal. The exclusive Kratix interface is the Kubernetes API, so any interaction with the tool is done through manipulating Kubernetes objects. You can overlay developer portals like Backstage onto both solutions to create a customized developer experience.

Score, a CNCF Sandbox project, is used by developers to specify workloads and their dependencies with the Humanitec Platform Orchestrator. The Score specification provides a curated set of features for containerized workloads. With Kratix, you compose your own high level offerings for different workload types yourself by combining and abstracting individual components.

An important distinction is that while all tools are able to provision a set of required resources, the Humanitec Platform Orchestrator is fully graph-based. Resources can interact with each other by querying the graph to obtain values required for a consistent configuration of the whole. The graph can be modified indirectly by the requirements developers set forth in their Score spec. This allows you to create powerful patterns applicable universally and at scale.

The Humanitec Platform Orchestrator comes with predefined RBAC and with turnkey solutions for secure communication and for managing your own secrets, fulfilling enterprise grade security posture requirements.

In most organizations it will not make sense to use the Humanitec products and another orchestration tooling side-by-side as this would multiply the interfaces for developers and the maintenance effort for platform engineers.