All Posts

Many organisations have successfully adopted GitOps for application delivery, but platform infrastructure provisioning often remains stubbornly manual, especially within enterprises. And by the end of this guide, you’ll see what this looks like in practice. You’ll request a Redis instance with a simple YAML manifest, Kratix will process the request, and Flux will automatically deploy the resources across your clusters.

Independent research reinforces this. MIT’s 2025 State of AI in Business report found that despite tens of billions in spend, only 5% of custom enterprise AI tools make it to production. Enterprises lead in pilot volume yet lag in scale. The findings point to approach, not model quality, as the real limiter. This article explores three platform-engineering principles that consistently turn AI from hype into working enterprise systems.

Workflow engines are having a moment in platform engineering. Tools like Temporal, Camunda, or n8n are being adopted to manage stateful automation, retries, and human approvals. They’re powerful, but they’re not platforms.  This post distils the workflow-engine thread from our webinar “Beyond the Platform Facade: Why Portals and Pipelines Aren’t Enough,” which Abby hosted and I presented.

Many organisations have invested heavily in Terraform, and rightly so. It’s a mature Infrastructure as Code (IaC) tool that automates infrastructure and streamlines workflows. However, when building an internal developer platform (IDP), IaC alone isn’t enough. To scale across teams and business needs, all while supporting seamless fleet management, you also need developer-friendly platform APIs and abstractions. That is where Kratix comes in.

Cloud-native engineering (and platform engineering ) is evolving quickly, and AI is at the centre of that shift. In the past, when a developer needed something as simple as a new database for testing, the process was slow. They had to raise a ticket, wait for approvals, and depend on manual provisioning, which sometimes took days before they could even start working.          Now picture the same request in 2025. Instead of waiting, an AI assistant provisions the database ins