CI/CD, GitOps, Kubernetes, and the platforms that ship software
DevOps started as a cultural movement to break down the wall between developers and operators. A decade later the practice has split, matured, and rebranded itself several times — site reliability engineering, platform engineering, DevSecOps, FinOps. The SoftwareMarketplace.Net DevOps hub publishes for the engineers actually doing the work: building pipelines that survive contact with reality, running Kubernetes clusters that do not page the on-call at 3 a.m., and shipping the internal developer platforms that make small product teams move fast without operating their own infrastructure.
Our coverage is organized around the four capabilities that matter most. Continuous delivery: pipeline architecture, secure supply chains, signed artifacts, progressive delivery patterns, and the metrics that let you trust your own deployments. GitOps and infrastructure as code: ArgoCD, Flux, Terraform, Pulumi, Crossplane, and the operational patterns that let declarative tooling scale beyond demos. Kubernetes operations: multi-cluster design, operator patterns, cost control, autoscaling, and the failure modes we have actually debugged in production. Platform engineering: golden paths, internal developer portals, paved roads versus paved gates, and how to staff a platform team that ships product engineers' productivity rather than its own roadmap.
We write for DevOps engineers, SREs, platform engineers, and the engineering managers who are trying to decide whether platform engineering is a real discipline or a rebrand. Every tutorial is tested against current tool versions. Every reference architecture is a working starting point, not a vendor whitepaper. Where we recommend a tool we explain the constraints under which the recommendation holds.
Start with the pillar guides if you are designing a new delivery pipeline or evaluating whether to invest in a platform team. Use the latest articles to keep up with releases that materially change how Kubernetes, Terraform, ArgoCD, or the major cloud control planes operate. For teams new to the discipline, the beginner guides build the foundation you need to be productive on a modern delivery team.
Both ArgoCD and Flux deliver the GitOps promise, but the operational shape of each tool is different. Here is how to choose between them.
Six patterns that separate CI/CD pipelines that survive a 10x increase in engineers from the ones that become a permanent platform-team backlog.
Operators turn operational knowledge into running code. Here are the patterns that hold up in production and the failure modes to design around.
The three pillars model is a useful starting point and a misleading destination. Here is what production observability actually looks like in 2026.
