Skip to main content

Platform Engineering Learning Path

Design and build Internal Developer Platforms (IDPs) that accelerate engineering teams. This is a senior track: plan ~12–16 weeks if the prerequisites are fresh, longer if you're building them here.

By the end you'll be able to: design a self-service platform on Kubernetes, codify golden paths with IaC and GitOps, enforce guardrails with policy-as-code, and make the product case for a platform team — the skills behind the Platform Engineer career path.

What is Platform Engineering?

Platform Engineering is the practice of building internal platforms that give development teams self-service capabilities — reducing cognitive load and improving developer experience. If DevOps is a culture, the platform is its product.

Stage 0: Prerequisites — learn them here

This path assumes solid ground in four areas, all covered on CloudCaptain:

PrerequisiteGuide
CI/CDFundamentals + GitOps
Infrastructure as CodeTerraform FundamentalsAdvanced
Containers & KubernetesDockerKubernetesWorkloads
One cloud providerAWS · Azure · GCP

Not there yet? Run the DevOps path and Containers path first.

Stage 1: Platform Foundations (2–3 weeks)

  • Understand developer workflows and pain points — the platform's "user research"
  • API design and service architecture — platforms are consumed programmatically
  • Infrastructure automation at scale — Terraform modules as reusable building blocks
  • Kubernetes production operations — the substrate most IDPs run on

Stage 2: Platform Building Blocks (3–4 weeks)

  • Service catalogs & developer portals — Backstage, Port, Cortex
  • Self-service infrastructure — Crossplane, Terraform modules behind an API
  • Golden paths — templated, standardized workflows from repo-creation to production
  • GitOps deliveryArgoCD/FluxCD as the deployment backbone
  • Internal CLIs — custom tooling; Bash and Python scripting pay off here

Stage 3: Advanced Topics (3–4 weeks)

  • Multi-tenancy and resource management on shared clusters
  • Policy as Code — OPA, Kyverno; guardrails instead of gates (see Kubernetes Security and DevSecOps)
  • FinOps integration — showback/chargeback per team
  • Platform metrics and ROI — DORA metrics, developer satisfaction surveys
  • Building and running platform teams

Key Principles

  1. Treat your platform as a product — your developers are your customers
  2. Self-service first — reduce tickets and waiting
  3. Golden paths, not golden cages — guide, don't restrict
  4. Measure everything — DORA metrics, developer satisfaction
  5. Iterate continuously — platforms are never "done"

Prove It

  1. Build — stand up a mini-IDP: Backstage + a Terraform module catalog + ArgoCD, letting a "developer" self-serve a new service to production. This is a portfolio-grade project.
  2. Interview — platform interviews lean on the fundamentals: Kubernetes · Terraform · CI/CD · DevOps
  3. Level up — the SRE path is the natural sibling track.