Sign inStart practicing
Start practicing Sign in

Platform Engineering and SRE Roadmap: Kubernetes to SLOs

A platform engineering and SRE roadmap: containers, Kubernetes, infrastructure as code, CI/CD, observability, SLOs, incident response and resilience.

Intermediate 3 min read · Updated

This platform engineering and SRE roadmap is for engineers interviewing for platform, infrastructure, DevOps and site reliability roles. It starts with how reliability is defined and budgeted, then covers containers and Kubernetes, infrastructure as code and delivery pipelines, service networking, observability, resilience patterns, and the people-and-process side of running production: incidents, capacity, multi-tenancy, secrets and internal developer platforms.

Platform design rounds are judged on safety as much as scale. For every component you add, say how it fails, how you'd detect that, and how you'd roll back.

Builds on the core system design roadmap. Do that first, or skip the parts you already know; this one only covers what platform engineering and sre interviews go deeper on.

Tick topics off as you read them

  1. Step 1SLOs and capacity

    How reliability is promised, budgeted and paid for.

  2. Step 2Compute and orchestration

    Packaging and scheduling workloads.

  3. Step 3Infrastructure and delivery

    Changing production safely and repeatably.

    Practice:Safe deployment platformPlatform · IntermediateConfiguration control planePlatform · Advanced

  4. Step 4Service-to-service networking

    How services find, trust and talk to each other.

    Practice:Service discovery for a growing fleetPlatform · Intermediate

  5. Step 5Observability at scale

    Storing and querying the signals a fleet produces.

    Practice:Metrics and observability platformPlatform · Advanced

  6. Step 6Resilience

    Surviving failures and overload, and proving it.

    Practice:Multi-region failover strategyPlatform · Advanced

  7. Step 7Operating at scale

    People, process and shared platforms.

    Practice:Secrets management platformPlatform · AdvancedInternal developer platformPlatform · Advanced · Free

What platform and SRE interviews test

AreaTypical promptGuides
Reliability targetsSet SLOs for a new API and decide what happens when the error budget runs out.Availability and SLOs, incident management
DeliveryDesign a deployment platform that lets 200 teams ship safely.CI/CD, deployment strategies, feature flags
InfrastructureManage cloud resources for many teams without drift or outages.Infrastructure as code, Kubernetes
ObservabilityDesign a metrics and alerting platform for a large fleet.Observability, time-series databases
ResilienceSurvive the loss of a region with a stated RPO and RTO.Disaster recovery, chaos engineering
Shared platformsBuild an internal developer platform or a secrets service.Developer platforms, secrets

A five-week plan

  1. Week 1: steps 1 and 2. Define SLIs and SLOs for a checkout API, then explain how a Kubernetes deployment rolls out a new version.
  2. Week 2: step 3. Design a pipeline from commit to production with a canary stage and automatic rollback; try the deployment platform round.
  3. Week 3: steps 4 and 5. Design service discovery for a growing fleet and the metrics behind its dashboards.
  4. Week 4: step 6. Design a multi-region failover strategy and plan a game day to test it.
  5. Week 5: step 7. Design a secrets platform and an internal developer platform.

Frequently asked questions

What is the difference between platform engineering, DevOps and SRE?

DevOps is a culture of developers and operators sharing ownership of production. SRE is Google's discipline for running production with engineering methods, built around SLOs and error budgets. Platform engineering builds the internal products (pipelines, templates, self-service infrastructure) that let product teams practise both without each reinventing them.

How much Kubernetes do I need to know?

Know the control plane and the reconciliation loop, the core objects (pods, deployments, services, config and secrets), how scheduling and autoscaling work, and the common failure modes. The Kubernetes architecture guide covers that level.

Do SRE interviews include system design?

Usually, alongside troubleshooting and coding rounds. The design round often focuses on monitoring, deployment safety or failure handling rather than product features.

Further reading