Rethinking DevOps: From Scripts to Platforms, From Infra as Code to Infra as Interfaces.

Table of contents
- ๐งฉ ๐ง๐ต๐ฒ ๐ฅ๐ฒ๐ฎ๐น ๐ฃ๐ฟ๐ผ๐ฏ๐น๐ฒ๐บ: ๐๐ฒ๐๐ข๐ฝ๐ itself ๐๐ผ๐ฒ๐๐ปโ๐ ๐ฆ๐ฐ๐ฎ๐น๐ฒ.
- ๐ ๐ง๐ต๐ฒ ๐ฆ๐ต๐ถ๐ณ๐: ๐๐ฟ๐ผ๐บ ๐๐ป๐ณ๐ฟ๐ฎ ๐ฎ๐ ๐๐ผ๐ฑ๐ฒ โ ๐๐ป๐ณ๐ฟ๐ฎ ๐ฎ๐ ๐๐ฃ๐๐
- ๐งฌ ๐๐ป๐๐ฒ๐ฟ: ๐๐ป๐๐ฒ๐ฟ๐ป๐ฎ๐น ๐๐ฒ๐๐ฒ๐น๐ผ๐ฝ๐ฒ๐ฟ ๐ฃ๐น๐ฎ๐๐ณ๐ผ๐ฟ๐บ๐ (๐๐๐ฃ๐)
- ๐ ๐ง๐ต๐ฒ ๐๐ฟ๐ฐ๐ต๐ถ๐๐ฒ๐ฐ๐๐๐ฟ๐ฒ ๐โ๐บ ๐๐๐ถ๐น๐ฑ๐ถ๐ป๐ด: ๐๐ป๐๐ฒ๐ฟ๐ณ๐ฎ๐ฐ๐ฒ๐ ๐ผ๐๐ฒ๐ฟ ๐ฆ๐ฐ๐ฟ๐ถ๐ฝ๐๐
- ๐ ๐๐ถ๐ป๐ฎ๐น ๐ง๐ต๐ผ๐๐ด๐ต๐
- The future of DevOps isn't more YAML โ it's ๐ฎ๐ฏ๐๐๐ฟ๐ฎ๐ฐ๐๐ถ๐ผ๐ป, ๐ฎ๐๐๐ผ๐บ๐ฎ๐๐ถ๐ผ๐ป, ๐ฎ๐ป๐ฑ ๐ฒ๐ ๐ฝ๐ฒ๐ฟ๐ถ๐ฒ๐ป๐ฐ๐ฒ. IDPs are how we get there. Kubernetes is how we standardize it. APIs are how we scale it.

Over the years, Iโve seen the ๐๐ก๐๐ ๐น๐ฎ๐ป๐ฑ๐๐ฐ๐ฎ๐ฝ๐ฒ and the broader open-source ecosystem evolve dramatically. It started with solving specific infrastructure challenges โ observability, container orchestration, security, service meshes, CI/CD, and so on. Each tool had its own purpose, and teams would hand pic the tools and stitch them together based on their needs.
But lately, Iโve started noticing a ๐๐ต๐ถ๐ณ๐ โ it doesnโt feel like itโs just about tools anymore. Thereโs a growing trend, especially in the ๐ฐ๐น๐ผ๐๐ฑ ๐ฎ๐ป๐ฑ ๐๐ฒ๐๐ข๐ฝ๐ ๐๐ฝ๐ฎ๐ฐ๐ฒ, toward building something more unified. Instead of just picking tools off the shelf, teams are starting to build ๐ฝ๐น๐ฎ๐๐ณ๐ผ๐ฟ๐บ๐โ composable, API-driven systems that developers can actually build on top of. Itโs less about operations reacting to needs, and more about ๐ฒ๐ป๐ด๐ถ๐ป๐ฒ๐ฒ๐ฟ๐ถ๐ป๐ด ๐ถ๐ป๐ณ๐ฟ๐ฎ๐๐๐ฟ๐๐ฐ๐๐๐ฟ๐ฒ ๐น๐ถ๐ธ๐ฒ ๐ฎ ๐ฝ๐ฟ๐ผ๐ฑ๐๐ฐ๐.
๐งฉ ๐ง๐ต๐ฒ ๐ฅ๐ฒ๐ฎ๐น ๐ฃ๐ฟ๐ผ๐ฏ๐น๐ฒ๐บ: ๐๐ฒ๐๐ข๐ฝ๐ itself ๐๐ผ๐ฒ๐๐ปโ๐ ๐ฆ๐ฐ๐ฎ๐น๐ฒ.
DevOps bridged the gap between developers and operations, but over time it became ๐๐ฐ๐ฟ๐ถ๐ฝ๐-๐ต๐ฒ๐ฎ๐๐, ๐ฌ๐๐ ๐-๐ฏ๐น๐ผ๐ฎ๐๐ฒ๐ฑ, and increasingly complex. As we scaled, we hit diminishing returns:
โข Terraform state files crack at team scale.
โข CI/CD pipelines became fragile glue
โข Ops turned into gatekeepers again
In one startup I worked with, developers had to wait days for infra tickets to be fulfilled. When we moved to a GitOps-driven system backed by Crossplane, deployment time and provisioning time dropped from ๐ฐ๐ฑ ๐บ๐ถ๐ป๐๐๐ฒ๐ ๐๐ผ ๐๐ป๐ฑ๐ฒ๐ฟ ๐ญ๐ฌ, and onboarding new services became self-service.
The takeaway?
What we need is not more automation. We need DevOps itself to scale โ through abstraction.
๐ ๐ง๐ต๐ฒ ๐ฆ๐ต๐ถ๐ณ๐: ๐๐ฟ๐ผ๐บ ๐๐ป๐ณ๐ฟ๐ฎ ๐ฎ๐ ๐๐ผ๐ฑ๐ฒ โ ๐๐ป๐ณ๐ฟ๐ฎ ๐ฎ๐ ๐๐ฃ๐๐
Kubernetes is no longer just a container orchestrator. Itโs becoming a ๐๐ป๐ถ๐๐ฒ๐ฟ๐๐ฎ๐น ๐ฐ๐ผ๐ป๐๐ฟ๐ผ๐น ๐ฝ๐น๐ฎ๐ป๐ฒ โ an interface layer for infrastructure, applications, workflows, security, policies etc.
With tools like ๐๐ฟ๐ผ๐๐๐ฝ๐น๐ฎ๐ป๐ฒ, we can define cloud resources as CRDs and let Kubernetesโ reconciliation engine manage state. Git becomes the source of truth, and the K8s API becomes the contract. No brittle scripts. No waiting.
This is not just โinfrastructure as code.โ
This is ๐ถ๐ป๐ณ๐ฟ๐ฎ๐๐๐ฟ๐๐ฐ๐๐๐ฟ๐ฒ ๐ฎ๐ ๐ฝ๐ฟ๐ผ๐ด๐ฟ๐ฎ๐บ๐บ๐ฎ๐ฏ๐น๐ฒ, ๐ฑ๐ฒ๐ฐ๐น๐ฎ๐ฟ๐ฎ๐๐ถ๐๐ฒ ๐๐ฃ๐๐.
๐งฌ ๐๐ป๐๐ฒ๐ฟ: ๐๐ป๐๐ฒ๐ฟ๐ป๐ฎ๐น ๐๐ฒ๐๐ฒ๐น๐ผ๐ฝ๐ฒ๐ฟ ๐ฃ๐น๐ฎ๐๐ณ๐ผ๐ฟ๐บ๐ (๐๐๐ฃ๐)
So where does this leave developer experience?
This is where ๐๐๐ฃ๐ come in โ and theyโre misunderstood. An IDP isnโt a dashboard for ops scripts. Itโs a ๐ฑ๐ฒ๐๐ฒ๐น๐ผ๐ฝ๐ฒ๐ฟ-๐ณ๐ฎ๐ฐ๐ถ๐ป๐ด ๐ฝ๐ฟ๐ผ๐ฑ๐๐ฐ๐ that exposes safe, self-service access to infrastructure and applications.
Tools like ๐๐ฎ๐ฐ๐ธ๐๐๐ฎ๐ด๐ฒ and ๐ฃ๐ผ๐ฟ๐ are not just UIs. They serve as:
โ Standardized and safe templates for infra and app scaffolding.
โ Secure abstraction layers over GitOps / Kubernetes APIs
โ A product mindset for internal tooling
An IDP isn't about hiding infrastructure โ it's about productizing it.
๐ ๐ง๐ต๐ฒ ๐๐ฟ๐ฐ๐ต๐ถ๐๐ฒ๐ฐ๐๐๐ฟ๐ฒ ๐โ๐บ ๐๐๐ถ๐น๐ฑ๐ถ๐ป๐ด: ๐๐ป๐๐ฒ๐ฟ๐ณ๐ฎ๐ฐ๐ฒ๐ ๐ผ๐๐ฒ๐ฟ ๐ฆ๐ฐ๐ฟ๐ถ๐ฝ๐๐
Iโm working on an Internal Developer Platform (IDP) that enables:
๐ Port or Backstage โ Git PR โ Argo CD or Flux โ Kubernetes โ Crossplane โ Infra provisioned
Hereโs how it works:
โข Developers scaffold services from templates (Port/Backstage)
โข App and infra manifests trigger a Git PR
โข GitOps tools deploy via Argo CD or Flux
โข Crossplane provisions infra dynamically
Kubernetes acts as a universal control plane โ APIs for apps, infra, workflows, and policies
โ Self-service
โ API-first
โ Auditable & scalable
This isn't just a toolchain โ it's a shift in mindset:
โข Interfaces > Scripts
โข Reconciliation > Imperative Pipelines
โข Platforms > Pipelines
โข Developer Experience > Infra Internals
Infrastructure is no longer something handed to developers โ itโs something they build and own, powered by APIs and self-service platforms.โ
And Kubernetes?
Itโs no longer just a container orchestrator โ itโs becoming the operating system for the cloud-native stack.
๐ ๐๐ถ๐ป๐ฎ๐น ๐ง๐ต๐ผ๐๐ด๐ต๐
The future of DevOps isn't more YAML โ it's ๐ฎ๐ฏ๐๐๐ฟ๐ฎ๐ฐ๐๐ถ๐ผ๐ป, ๐ฎ๐๐๐ผ๐บ๐ฎ๐๐ถ๐ผ๐ป, ๐ฎ๐ป๐ฑ ๐ฒ๐ ๐ฝ๐ฒ๐ฟ๐ถ๐ฒ๐ป๐ฐ๐ฒ. IDPs are how we get there. Kubernetes is how we standardize it. APIs are how we scale it.
Pipelines ship code. Platforms scale teams.
#PlatformEngineering #Kubernetes #CloudNative #InfrastructureAsAPI #Crossplane #InternalDeveloperPlatform #SystemDesign #DeveloperExperience
Subscribe to my newsletter
Read articles from Mohammad Ayan directly inside your inbox. Subscribe to the newsletter, and don't miss out.
Written by

Mohammad Ayan
Mohammad Ayan
I am a B.Tech graduate and DevOps Engineer with 2.5+ years of experience in cloud and DevOps field. I hold the ๐๐๐ซ๐ญ๐ข๐๐ข๐๐ ๐๐ฎ๐๐๐ซ๐ง๐๐ญ๐๐ฌ ๐๐๐ฆ๐ข๐ง๐ข๐ฌ๐ญ๐ซ๐๐ญ๐จ๐ซ (๐๐๐) and ๐๐๐ ๐๐๐ซ๐ญ๐ข๐๐ข๐๐ ๐๐ฅ๐จ๐ฎ๐ ๐๐ซ๐๐๐ญ๐ข๐ญ๐ข๐จ๐ง๐๐ซ certifications. My expertise spans multiple cloud platforms, including ๐๐๐ and ๐๐๐, where I excel at implementing best practices to enhance security, automate the SDLC for faster rollouts, and improve efficiency while minimizing friction between development and production. I am also skilled in implementing observability solutions at both the infrastructure and application levels, ensuring early issue detection and monitoring through robust notification systems. In my roles as a Cloud Engineer and DevOps Engineer, I have worked on multiple projects for various companies. I implemented ๐๐/๐๐ ๐ฉ๐ข๐ฉ๐๐ฅ๐ข๐ง๐๐ฌ to facilitate faster rollouts, leveraged ๐๐ฎ๐๐๐ซ๐ง๐๐ญ๐๐ฌ for deployments, and managed AWS and GCP organization accounts with Service Control Policies (SCP) enforced. My experience with observability solutions, including ๐๐ซ๐จ๐ฆ๐๐ญ๐ก๐๐ฎ๐ฌ,๐๐ซ๐๐๐๐ง๐, and the ELK stack, ensures robust monitoring and early issue detection. Additionally, I have crafted ๐ฅ๐จ๐ฐ-๐ฅ๐๐ญ๐๐ง๐๐ฒ, ๐๐จ๐ฌ๐ญ-๐จ๐ฉ๐ญ๐ข๐ฆ๐ข๐ฌ๐๐, and ๐ก๐ข๐ ๐ก๐ฅ๐ฒ ๐๐ฏ๐๐ข๐ฅ๐๐๐ฅ๐ solutions, supported by monitoring systems and dashboards for complete observability. I enjoy writing about technologies and trends in the cloud-native space, discussing issues and sharing insights through my blogs on Medium and Hashnode. I aim to contribute to the tech community by providing valuable content for developers and engineers. You can also check out my portfolio website for a deeper look at my work and projects. https://muhmmadayanashiq.wixsite.com/ayan-portfolio-1 (๐๐จ๐ซ๐ญ๐๐จ๐ฅ๐ข๐จ) https://www.credly.com/badges/448a7101-d397-4a57-a30d-5ee02a2dab59/public_url (๐๐๐ ๐๐๐ซ๐ญ๐ข๐๐ข๐๐๐ญ๐) https://muhmmadayan.hashnode.dev/ (๐๐๐ฌ๐ก๐ง๐จ๐๐ ๐๐ฅ๐จ๐ ) muhmmadayanashiq@gmail.com (๐๐ฆ๐๐ข๐ฅ ๐๐๐๐ซ๐๐ฌ๐ฌ)