> Markdown version of [/videos/100137-building-a-cloud-platform-where-everything-is-just-another-kubernetes-resource](https://www.wearedevelopers.com/videos/100137-building-a-cloud-platform-where-everything-is-just-another-kubernetes-resource). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md).

---

# Building a Cloud Platform Where Everything is Just Another Kubernetes Resource

Scaling Terraform caused grueling ten-hour deployment plans. Learn how shifting to a Kubernetes control plane eliminates infrastructure drift and supercharges self-service provisioning.

- **Speakers:** [Patrick Koss](https://www.wearedevelopers.com/@patrick-koss)
- **Event:** World Congress 2026 Europe
- **Published:** July 9, 2026
- **Duration:** 28:49
- **URL:** https://www.wearedevelopers.com/videos/100137-building-a-cloud-platform-where-everything-is-just-another-kubernetes-resource

## Summary

By transitioning from traditional infrastructure-as-code paradigms to a Kubernetes control plane model, tech teams can drastically reduce operational complexity and provisioning bottlenecks. At Stackit Cloud, scaling with Terraform eventually introduced painful trade-offs: massive serialized plans taking up to 10 hours, cumbersome failure recoveries, rigid deletion safety nets, and the mental overhead of maintaining dual paradigms alongside their native Kubernetes workloads.

To solve this, the engineering team rebuilt their entire cloud interaction layer around Kubernetes Custom Resource Definitions (CRDs), utilizing Crossplane and a GitOps pipeline driven by ArgoCD. In this model, every component—from object storage and databases to DNS records and credentials—is defined as a simple YAML manifest. Rather than manual point-in-time applies, Kubernetes controllers continuously reconcile the desired state against the actual cloud environment, eliminating unauthorized drift. To ensure the central Kubernetes API doesn't become overloaded, a hub-and-spoke architecture utilizes a core control plane cluster that selectively fans out instructions to distinct data plane clusters.

This API-driven architecture ultimately acts as a powerful self-service platform engine. Internal engineering teams simply apply Kubernetes resources to provision holistic environments—complete with observability endpoints, cert-managers, and backup storage—without deep operational knowledge. Crucially, platform operators can natively enforce strict governance policies using admission webhooks like Kyverno. By intercepting requests before they are committed, organizations can automatically require security fundamentals, such as blocking public internet access for storage buckets or mandating multi-AZ redundancy for production databases, preventing catastrophic "oopsie daisies" before they happen.

**Keywords:** kubernetes control plane, terraform scaling challenges, crossplane infrastructure provisioning, continuous state reconciliation, gitops pipeline deployment, argocd continuous delivery, self-service developer platforms, kubernetes custom resource definitions, admission webhook governance, kyverno policy enforcement, kubernetes api fan-out architecture, infrastructure drift remediation, declarative yaml manifests, platform engineering operations, cloud resource lifecycle management

## Chapters

1. **Managing cloud infrastructure with Kubernetes custom resources** (00:14) — Mapping YAML manifests directly to REST API calls simplifies cloud resource provisioning and reduces context switching.
1. **Limitations of infrastructure as code at massive scale** (04:06) — Monolithic state files, serialized pipeline executions, and fragile deletion policies create unmanageable bottlenecks in complex environments.
1. **Comparing static applies with continuous Kubernetes reconcile loops** (08:51) — Eventual consistency and continuous reconciliation offer significant reliability advantages over static state files and manual pipelines.
1. **Provisioning cloud infrastructure using Crossplane custom controllers** (12:50) — Implementing custom controllers that rely on long polling and simplified interfaces allows seamless management of external components.
1. **Designing a self-service internal developer platform with GitOps** (18:37) — Bootstrapping standardized environments using self-service APIs and distributed secret management drastically accelerates secure application deployment.
1. **Scaling control planes and enforcing governance via admission webhooks** (23:41) — Distributing workloads across fan-out architectures prevents API server overload while policy engines enforce strict security rules.

## Related Moments

- [Provisioning component dependencies using Crossplane and Kubernetes manifests](https://www.wearedevelopers.com/videos/1197-dapr-in-practice) (from "Dapr in Practice")
- [Utilizing Kubernetes as a foundation for internal platforms](https://www.wearedevelopers.com/videos/186-effective-java-strategies-and-architectures-for-clouds) (from "Effective Java Strategies and Architectures for Clouds")
- [Demonstrating rapid application deployments with Stackit cloud foundry](https://www.wearedevelopers.com/videos/413-the-future-of-cloud-is-abstraction-why-kubernetes-is-not-the-endgame-for-stackit) (from "The Future of Cloud is Abstraction - Why Kubernetes is not the Endgame for STACKIT ")
- [Architecture of the cloud native web application platform](https://www.wearedevelopers.com/videos/1519-empowering-thousands-of-developers-our-journey-to-an-internal-developer-platform) (from "Empowering Thousands of Developers: Our Journey to an Internal Developer Platform")
- [Extending the Kubernetes control plane with custom resources](https://www.wearedevelopers.com/videos/1065-chaos-in-containers-unleashing-resilience) (from "Chaos in Containers - Unleashing Resilience")
- [Managing cluster platform infrastructure via GitOps principles](https://www.wearedevelopers.com/videos/1582-llmops-driven-fine-tuning-evaluation-and-inference-with-nvidia-nim-nemo-microservices) (from "LLMOps-driven fine-tuning, evaluation, and inference with NVIDIA NIM & NeMo Microservices")

## Related Articles

- [Learning Kubernetes made easy with KubeCampus](https://www.wearedevelopers.com/magazine/348-learning-kubernetes-made-easy-with-kubecampus)
- [How we Build The Software of Tomorrow](https://www.wearedevelopers.com/magazine/120-how-we-build-the-software-of-tomorrow)
- [Never delegate the understanding](https://www.wearedevelopers.com/magazine/749-never-delegate-the-understanding)
- [Now is the time for industrialized software development](https://www.wearedevelopers.com/magazine/601-now-is-the-time-for-industrialized-software-development)

## Related Jobs

- [Cloud Foundations Team](https://www.wearedevelopers.com/jobs/ext/1483289-cloud-foundations-team) at **GitHub**
- [Platform Engineer (DevOps)](https://www.wearedevelopers.com/jobs/48264-platform-engineer-devops) at **WDW Consulting GmbH**
- [Senior Cloud Native Solution Architect (all genders welcome) - Kubernetes, CNCF, MlOps](https://www.wearedevelopers.com/jobs/ext/101488-senior-cloud-native-solution-architect-all-genders-welcome-kubernetes-cncf-mlops) at **Rosenxt Group**
- [Cloud-Native Architect (all genders welcome) - Kubernetes, CNCF, MLOps](https://www.wearedevelopers.com/jobs/ext/101479-cloud-native-architect-all-genders-welcome-kubernetes-cncf-mlops) at **Rosenxt Group**
- [Lead Cloud DevSecOps Engineer - Kubernetes](https://www.wearedevelopers.com/jobs/ext/1659167-lead-cloud-devsecops-engineer-kubernetes) at **BWI GmbH**
- [Senior Cloud Native Solution Architect (all genders welcome) - Kubernetes, CNCF, MlOps](https://www.wearedevelopers.com/jobs/ext/66342-senior-cloud-native-solution-architect-all-genders-welcome-kubernetes-cncf-mlops) at **Rosenxt Group**