Skip to content

CNPA Study Plan

Five weeks at 4-6 hours per week. This is a concepts exam; reading and reasoning matter more than lab time.

Week 1: The white paper and platform fundamentals

  • Read the CNCF Platforms White Paper end to end
  • Define, in your own words: platform, product, golden path, cognitive load, thinnest viable platform
  • Platform as a product: users, value, roadmap, feedback, voluntary adoption
  • Why mandating adoption is an anti-pattern
  • Team Topologies vocabulary: platform team, enabling team, stream-aligned team, cognitive load
  • Interfaces and abstractions: what to hide, what to expose, leaky abstractions
  • Review Notes: notes/01-platform-fundamentals.md

Week 2: Maturity model and platform boundaries

  • Read the Platform Engineering Maturity Model
  • The levels: provisional, operational, scalable, optimizing
  • The attributes: investment, adoption, interfaces, operations, measurement
  • Place a platform you know at a level for each attribute, and justify it
  • Anti-patterns: no users, mandated adoption, rebuilding managed services, ticket-queue platform teams
  • Where a platform should stop, and when to buy rather than build

Week 3: Observability, security, and conformance

  • Platform observability versus observability offered to tenants
  • SLOs for platform capabilities and error budgets
  • Multi-tenancy models: namespace, cluster, and control plane isolation, with trade-offs
  • Tenant isolation controls: RBAC, network policy, quotas, admission control, runtime isolation
  • Policy as code with Kyverno and OPA Gatekeeper
  • Supply chain: signing, provenance, SBOMs, admission verification
  • Conformance validation and compliance evidence as a platform output
  • Review Notes: notes/02-observability-security-conformance.md

Week 4: Delivery, APIs, and provisioning

  • CI and CD boundaries in a platform context
  • GitOps as the delivery model; who owns what between platform and tenant
  • Progressive delivery and preview environments as self-service capabilities
  • The Kubernetes API as a platform API; custom resources as the interface
  • Operators and controllers as the reconciliation mechanism
  • Crossplane and control plane patterns; composition and claims
  • API design: versioning, deprecation, defaults, validation
  • Review Notes: notes/03-delivery-apis-provisioning.md

Week 5: Developer experience, measurement, and review

  • Internal Developer Platform versus Internal Developer Portal
  • Backstage: catalog, software templates, TechDocs, scaffolding
  • Onboarding and time to first deployment
  • DORA metrics, and what each one tells you
  • SPACE framework dimensions
  • Adoption, satisfaction, cost, showback and chargeback
  • Vanity metrics versus actionable ones
  • Review Notes: notes/04-developer-experience-and-measurement.md
  • Work every scenario in scenarios.md
  • Two timed practice exams

Readiness check

  • Explain what makes something a platform rather than a collection of tools
  • Define a golden path and say why it must not be the only path
  • Argue why mandating adoption hides failure
  • Place a platform on the maturity model across all attributes
  • Compare namespace, cluster, and control plane multi-tenancy with a use case for each
  • Explain the difference between an Internal Developer Platform and a portal
  • Name the four DORA metrics and what each reveals
  • Explain what a Crossplane composition and claim do