DevOps From Zero
Your DevOps Roadmap From Zero to Production
A practical path for complete beginners.
No previous tech experience required.
Don't learn random tools. Learn how real systems get built, shipped, operated, and recovered.
No account or email required.
The clickable path
Eight stages. One production mindset.
Open each stage, then tap a concept for a concise what-it-is and why-you-need-it note. Follow the stages in order.
Stage 01
Foundations
Build the operating-system, network, and source-control instincts every later stage needs.Linux · Networking · Git/GitHub · Bash · PythonLinux
What it isThe operating system behind most cloud servers and containers.
Why you need itLearn files, permissions, processes, packages, services, and troubleshooting.
Official documentationNetworking
What it isLearn how systems communicate through IPs, ports, DNS, TCP, and HTTP.
Why you need itA huge number of production problems eventually become networking problems.
Git
What it isTracks code history and changes.
Why you need itBranches and pull requests let teams review work safely before merging.
Official documentationGitHub
What it isHosts Git repositories and engineering workflows.
Why you need itIt commonly becomes the trigger point for CI/CD automation.
Official documentationBash
What it isThe Linux shell used for commands and automation.
Why you need itLearn pipes, variables, exit codes, grep, curl, and simple scripts.
Official documentationPython
What it isUseful for automation, APIs, and operational tooling.
Why you need itBeginners only need enough Python initially to automate real engineering tasks.
Official documentationStage 02
Cloud
Learn cloud concepts first, then map them across AWS, Azure, and GCP.AWS · Azure · GCP · IAM · Compute · Storage · Networking · DNS · SecretsIAM / RBAC
What it isControls who or what can access cloud resources.
Why you need itLearn identities, roles, and least privilege before giving broad access.
Compute
What it isRuns applications on virtual machines or managed compute.
Why you need itChoose based on how much infrastructure you want to operate yourself.
Object storage
What it isStores files, backups, artifacts, and datasets as objects.
Why you need itIt is different from a normal server disk or relational database.
Virtual network
What it isDefines private networking, subnets, routes, and connectivity.
Why you need itUnderstand traffic paths before deploying services.
DNS
What it isMaps human-friendly names to infrastructure endpoints.
Why you need itDNS changes can directly affect production traffic.
Secrets
What it isStores passwords, API keys, and sensitive configuration outside source code.
Why you need itApplications should access secrets with tightly scoped identities.
Managed containers
What it isRuns containerized apps without managing Kubernetes directly.
Why you need itIt is often a better first choice when Kubernetes is unnecessary.
Managed Kubernetes
What it isAWS EKS, Azure AKS, and GCP GKE manage Kubernetes control planes.
Why you need itYou still own workload security, scaling, observability, and reliability.
Concept first. Provider second.
AWS, Azure, and GCP service map
Learn the job each service performs, then recognize its name in every cloud.
IAM / RBAC
What it isControls who or what can access cloud resources.
Why you need itLearn identities, roles, and least privilege before giving broad access.
Compute
What it isRuns applications on virtual machines or managed compute.
Why you need itChoose based on how much infrastructure you want to operate yourself.
Object storage
What it isStores files, backups, artifacts, and datasets as objects.
Why you need itIt is different from a normal server disk or relational database.
Virtual network
What it isDefines private networking, subnets, routes, and connectivity.
Why you need itUnderstand traffic paths before deploying services.
DNS
What it isMaps human-friendly names to infrastructure endpoints.
Why you need itDNS changes can directly affect production traffic.
Secrets
What it isStores passwords, API keys, and sensitive configuration outside source code.
Why you need itApplications should access secrets with tightly scoped identities.
Container registry
What it isStores versioned container images for deployment.
Why you need itUse an authenticated registry to scan, control, and promote immutable artifacts.
Managed containers
What it isRuns containerized apps without managing Kubernetes directly.
Why you need itIt is often a better first choice when Kubernetes is unnecessary.
Managed Kubernetes
Relational database
What it isStores structured data with relationships and transactional guarantees.
Why you need itManaged databases reduce operational work while you retain responsibility for data design and access.
Monitoring
What it isCollects operational signals from cloud resources and applications.
Why you need itUse it to detect unhealthy behavior, investigate incidents, and measure reliability.
Stage 03
Containers
Package applications into repeatable artifacts that behave consistently across environments.Docker · Images · Containers · Volumes · Networking · RegistriesDocker
What it isA platform for building, running, and distributing containers.
Why you need itIt gives teams a repeatable way to package application dependencies.
Official documentationDockerfile
What it isA versioned recipe for building a container image.
Why you need itClear, minimal instructions make builds repeatable and reviewable.
Official documentationImage
What it isA packaged, immutable version of your application.
Why you need itBuild once, scan it, store it in a registry, and deploy the same artifact.
Container
What it isA running instance of an image.
Why you need itContainers should be disposable; persistent data belongs outside them.
Registry
What it isStores container images.
Why you need itExamples include ECR, Azure Container Registry, and Google Artifact Registry.
Volume
What it isPersistent storage mounted into a container.
Why you need itUse volumes when data must survive container replacement.
Container networking
What it isConnects containers to each other and to external systems.
Why you need itLearn ports, DNS, bridge networks, and explicit exposure.
Multi-stage builds
What it isUses separate build stages inside one Dockerfile.
Why you need itIt keeps build tools out of the smaller production image.
Official documentationStage 04
Infrastructure as Code
Describe infrastructure in reviewed code instead of clicking it into existence.Terraform · State · Modules · Plan · Apply · DriftTerraform
What it isDeclares infrastructure using versioned configuration.
Why you need itIt makes environments reviewable and reproducible.
Official documentationProviders
What it isPlugins that connect Terraform to cloud and service APIs.
Why you need itPin versions and upgrade them deliberately.
Official documentationResources
What it isBlocks representing real infrastructure objects.
Why you need itThey describe the desired state of networks, services, databases, and more.
Variables
What it isTyped inputs that make configuration reusable.
Why you need itUse them for reviewed differences between environments.
Outputs
What it isNamed values exposed after Terraform evaluates configuration.
Why you need itThey connect modules and publish useful non-secret deployment information.
State
What it isTerraform's record of which real resources belong to your configuration.
Why you need itProtect remote state because it can contain sensitive infrastructure data.
Remote state
What it isStores Terraform state in a shared protected backend.
Why you need itIt enables collaboration, locking, backups, and controlled access.
Modules
What it isReusable groups of Terraform resources.
Why you need itUse them to express stable patterns without copying configuration.
Plan
What it isShows what Terraform wants to add, change, or destroy.
Why you need itReview the plan before applying anything to production.
Apply
What it isExecutes an approved Terraform plan.
Why you need itUse saved plans and guarded automation for production changes.
Drift
What it isWhen real infrastructure no longer matches declared configuration.
Why you need itA fresh Terraform plan helps detect it.
Stage 05
Delivery
Move reviewed code to production through repeatable tests, scans, releases, and recovery.CI/CD · Testing · Security scanning · Deployment strategies · RollbackCI
What it isAutomatically validates code before deployment.
Why you need itIt runs tests, linting, builds, and security checks after code changes.
CD
What it isAutomates delivery or deployment of validated changes.
Why you need itIt reduces manual steps while keeping approval boundaries explicit.
Pipeline
What it isAn ordered workflow from source change to verified release.
Why you need itEvery stage needs clear inputs, outputs, and failure behavior.
Build artifact
What it isAn immutable output produced by a build.
Why you need itPromote the same tested artifact instead of rebuilding per environment.
Unit tests
What it isChecks small pieces of behavior in isolation.
Why you need itThey provide fast feedback on deterministic logic.
Integration tests
What it isChecks components working together across real boundaries.
Why you need itThey catch database, network, and configuration problems unit tests cannot.
Dependency scanning
What it isChecks third-party packages for known vulnerabilities.
Why you need itKeep the gate current and fail on disallowed risk.
Container scanning
What it isChecks the final image and installed packages for vulnerabilities.
Why you need itScan the deployable artifact, not only source dependencies.
Deployment
What it isReleases a known artifact into an environment.
Why you need itA good deployment is observable, repeatable, and verifiable.
Rolling deployment
What it isReplaces instances gradually with a new version.
Why you need itIt limits disruption while healthy capacity keeps serving.
Blue/green
What it isRuns old and new environments side by side.
Why you need itTraffic can switch quickly, giving rollback a known-good destination.
Canary
What it isExposes a small amount of traffic to a new version first.
Why you need itUse real signals to decide whether to continue or stop.
Rollback
What it isReturns production to a previously known-good version.
Why you need itA production deployment should always have a recovery path.
Stage 06
Orchestration
Operate containerized workloads beyond a single machine with Kubernetes.Pods · Deployments · Services · Ingress · ConfigMaps/Secrets · Storage · RBAC · Helm · Autoscaling · Health probesDo not start here. Learn Kubernetes after you understand what it is orchestrating.
Pod
What it isThe smallest deployable Kubernetes unit.
Why you need itIt runs closely related containers sharing networking and lifecycle.
Deployment
What it isManages stateless Pods.
Why you need itIt maintains replicas and handles rolling updates and rollback.
ReplicaSet
What it isKeeps the requested number of Pods running.
Why you need itIt is usually controlled automatically by a Deployment.
Service
What it isProvides a stable endpoint for changing Pods.
Why you need itApplications use it instead of individual Pod addresses.
Ingress
What it isRoutes external HTTP and HTTPS traffic to Services.
Why you need itIt commonly handles hostnames, paths, and TLS entry into the cluster.
ConfigMap
What it isStores non-secret configuration.
Why you need itPods consume values through environment variables or files.
Secret
What it isStores sensitive Kubernetes configuration.
Why you need itStrong RBAC and often an external secrets manager are still required.
Namespace
What it isCreates logical separation inside a cluster.
Why you need itIt often separates teams, applications, or environments.
RBAC
What it isControls which identities can perform Kubernetes API actions.
Why you need itFollow least privilege and avoid unnecessary cluster-admin access.
PV
What it isRepresents persistent storage available to Kubernetes.
Why you need itIt exists independently from an individual Pod lifecycle.
PVC
What it isA workload's request for persistent storage.
Why you need itKubernetes binds the request to suitable persistent capacity.
StatefulSet
What it isRuns workloads needing stable identity or storage relationships.
Why you need itUse it when replicas cannot be treated as interchangeable.
Job
What it isRuns work that should finish and exit.
Why you need itIt is useful for batch processing and one-time operational tasks.
CronJob
What it isSchedules Jobs on a recurring schedule.
Why you need itUse it for bounded maintenance and batch work.
Liveness probe
What it isAnswers: should Kubernetes restart this container?
Why you need itUse it to recover from a process that is irreparably stuck.
Readiness probe
What it isAnswers: should this Pod receive traffic right now?
Why you need itIt removes unhealthy or warming Pods from Service traffic.
Startup probe
What it isProtects slow-starting applications from being killed too early.
Why you need itIt delays liveness checks until startup succeeds.
HPA
What it isAutomatically changes Pod replica counts based on metrics.
Why you need itPair it with sensible requests, limits, and capacity planning.
Requests / limits
What it isRequests reserve resources; limits cap consumption.
Why you need itBad values cause waste, throttling, or unstable workloads.
Helm
What it isPackages Kubernetes manifests into configurable reusable charts.
Why you need itIt helps teams version and install related resources together.
Official documentationStage 07
Observability
Use telemetry to understand what a system is doing and why it changed.Logs · Metrics · Traces · OpenTelemetry · Prometheus · Grafana · AlertsLogs
What it isRecords events and context from applications and infrastructure.
Why you need itUse them to understand exactly what happened.
Metrics
What it isNumeric measurements over time.
Why you need itThey reveal trends, saturation, latency, and errors.
Traces
What it isFollow one request across multiple services.
Why you need itThey show where distributed applications become slow or fail.
OpenTelemetry
What it isA vendor-neutral standard for generating telemetry.
Why you need itIt supports traces, metrics, and logs across modern systems.
Official documentationPrometheus
What it isCollects and queries time-series metrics.
Why you need itIt is widely used for service and Kubernetes monitoring.
Official documentationGrafana
What it isBuilds dashboards from telemetry data sources.
Why you need itUse focused views to make operational signals understandable.
Official documentationAlerts
What it isNotifies people or automation when a meaningful condition occurs.
Why you need itAlerts should be actionable and tied to user impact.
SLIs
What it isMeasurements of service behavior such as availability or latency.
Why you need itThey turn reliability into observable evidence.
SLOs
What it isTargets for acceptable service behavior over time.
Why you need itThey guide reliability decisions and error-budget tradeoffs.
You can answer
- What broke?
- Where?
- Why?
- What changed?
Stage 08
Modern Operations
Design reliable, secure, cost-aware operations where automation remains constrained and verifiable.Reliability · Incidents · Security · Cost · Automation · AI-assisted operationsReliability
What it isDesigns systems to provide dependable service under real conditions.
Why you need itUse objectives, failure analysis, and recovery engineering to manage risk.
Incident response
What it isCoordinates detection, triage, recovery, and learning during failures.
Why you need itClear roles and evidence reduce confusion when pressure is high.
Capacity
What it isPlans resources for expected demand and failure scenarios.
Why you need itGood capacity avoids saturation and unnecessary cost.
Cost engineering
What it isConnects architecture choices to measurable cloud spend.
Why you need itUnderstand unit cost, waste, and scaling tradeoffs.
Security
What it isProtects identities, software, data, and infrastructure.
Why you need itTreat least privilege, patching, and secure defaults as operational work.
Automation
What it isTurns repeatable operational work into controlled software.
Why you need itAutomate only when inputs, permissions, failures, and recovery are understood.
AI-assisted operations
What it isUses AI to investigate signals and propose operational actions.
Why you need itEngineers design context, permissions, verification, and recovery.
Agent permissions
What it isDefines exactly what an operational agent can read or change.
Why you need itDefault to scoped, time-bounded access and explicit escalation.
Evaluation
What it isTests whether AI-assisted decisions are accurate and safe.
Why you need itUse realistic cases, failure thresholds, and regression checks.
Guardrails
What it isConstrains unsafe inputs, outputs, and actions.
Why you need itGuardrails complement evaluation; they do not replace system design.
Human approval
What it isKeeps consequential actions behind an accountable decision point.
Why you need itShow evidence, impact, rollback, and uncertainty.
2027 operational loop
- Detect
- Investigate
- Correlate
- Recommend
- Approve
- Automate
- Verify
AI increasingly handles investigation and proposes operational actions. Engineers design the context, permissions, controls, verification, and recovery.
The engineer you're becoming
Tools are not the destination.
Not: someone who knows Docker, Terraform, and Kubernetes.
Instead: someone who can take a system from code to production, understand why it fails, recover it safely, and automate what should not require a human.
- Build
- Ship
- Operate
- Observe
- Recover
- Automate
Don't just learn it. Prove it.
Build proof that your skills work together.
Tutorials show you how. Open each project to see the proof a real portfolio should contain.
Project 01Deploy
Containerize and deploy a real API.
Proof requirements
- Working deployment
- Health endpoint
- Environment configuration
- Documented architecture
Project 02Automate
Provision the application's infrastructure with Terraform.
Proof requirements
- Reproducible environment
- Reviewed Terraform plan
- Remote state
- No manual infrastructure dependency
Project 03Ship
Build a production CI/CD pipeline.
- Commit
- Test
- Scan
- Build
- Deploy
- Verify
Proof requirements
- Automated quality gates
- Immutable release
- Verified deployment
- Rollback behavior
Project 04Scale
Run the application using orchestration.
Proof requirements
- Service discovery
- Configuration
- Secrets
- Scaling
- Rollout and recovery
Project 05Observe
Instrument the system.
Proof requirements
- Logs
- Metrics
- Traces
- Dashboards
- One actionable alert
Project 06Break + Recover
Create and recover from a controlled production-style failure.
Controlled failure examples
- Bad deployment
- CPU saturation
- Memory pressure
- Unavailable dependency
- Connectivity failure
- Consumer backlog
- Detect
- Diagnose
- Identify root cause
- Recover
- Document prevention
Proof requirements
- Evidence-led diagnosis
- Safe recovery
- Root-cause record
- Prevention plan
Learn with structure
Want me to teach you DevOps from zero?
I'm building DevOps From Zero for people who want a structured path from complete beginner to building real production projects.
The roadmap tells you what to learn. The program will teach you how to build it.
Joining the list does not guarantee enrollment, employment, salary, certification, or any particular outcome.