18 stations on 3 ladders, in order — each rung is written assuming the one above it. Mark them off as you go; your progress is saved in this browser.
0 / 18
Pick up at
Track 1 of 3
FoundationsStart here
6 rungs · ~5h · ends at Scripting: Bash & Python.
0/6 done
1
Linux & the Shell
Start here
F1 · Beginner · 60 min
Almost every server runs Linux, so DevOps starts at the command line. Learn the filesystem, permissions, processes, package managers, and enough shell to move confidently. The terminal is not a fallback here — it is the primary interface to everything you operate.
Filesystem & permissions
Processes & signals
Package managers
Pipes & redirection
Build itGiven a box you have never seen, find what is eating CPU, which process owns port 8080, and how much disk is left — all from the shell.
✓ CheckpointExplain what a signal does to a process, and why kill -9 is the one your service never gets to handle.
2
Networking Basics
F2 · Beginner · 60 min
You cannot operate systems you cannot trace. Learn IP and ports, DNS, TCP vs UDP, HTTP, TLS, and firewalls — enough to answer "why can this box not reach that one?" Networking is the single most common source of production mysteries.
Git is the substrate of DevOps — code, config, and increasingly infrastructure all live in it. Learn branching, merging, rebasing, and pull-request workflows, because in a GitOps world a git commit is how you change production. Master it beyond the three commands you know.
Branching & merging
Rebase vs merge
PR workflows
Tags & history
Build itTwo branches touched the same config. Resolve the conflict, then explain when you would rebase instead of merge — and when you must not.
✓ CheckpointExplain when rebase is the right tool and when it rewrites history someone else has already pulled.
4
Docker & Containers
F4 · Intermediate · 75 min
Containers package an app with its dependencies so it runs the same everywhere — the unit modern deployment is built on. Learn images, layers, the Dockerfile, registries, and why containers are not VMs. This is the gateway to everything orchestrated.
Images & layers
Dockerfiles
Registries
Containers vs VMs
Build itContainerize an app in a multi-stage Dockerfile so the final image is small. Why does layer order decide your build cache hit rate?
✓ CheckpointExplain why Docker layer order changes build time, and which line in your Dockerfile invalidates the cache most often.
DevOps runs on the cloud. Learn the core primitives every provider shares — compute, storage, networking (VPC), identity (IAM), and the managed-service tradeoff. The provider changes; the mental model — pay for what you use, everything is an API — does not.
Compute / storage / VPC
IAM & least privilege
Managed vs self-hosted
Regions & availability zones
Build itDesign the minimal cloud footprint for a web app: which pieces are managed services, and where does IAM draw the security boundary?
✓ CheckpointExplain the shared responsibility line for a managed service you use, and name what you wrongly assumed the provider handles.
6
Scripting: Bash & Python
F6 · Intermediate · 55 min
DevOps is automation, and automation is code. Learn enough Bash to glue commands and enough Python to write real tooling — parsing, calling APIs, handling errors. The rule: if you did it twice by hand, script it the third time.
Bash glue & exit codes
Python for tooling
Calling APIs
Idempotent scripts
Build itWrite a script that rotates old log files and alerts if disk crosses 80%. Make it safe to re-run — idempotent, no double action.
✓ CheckpointExplain what makes a script idempotent, and why that matters more than whether it is short.
Track 2 of 3
Core DevOpsThe craft
6 rungs · ~5h · ends at Artifacts & Registries.
0/6 done
1
CI/CD Pipelines
T1 · Intermediate · 65 min
Continuous integration and delivery turn every commit into a tested, deployable artifact automatically. Learn pipeline stages (build, test, scan, deploy), caching, parallelism, and gates. A good pipeline makes shipping boring — which is exactly the point.
Pipeline stages
Build caching & parallelism
Quality gates
Artifacts between stages
Build itDesign a pipeline that lints, tests, builds an image, scans it, and deploys to staging — and blocks promotion on any failed gate.
✓ CheckpointExplain what belongs in a quality gate that blocks a merge, and what belongs in a job that only reports.
2
Infrastructure as Code
T2 · Advanced · 70 min
Click-ops does not scale or reproduce. Infrastructure as Code (Terraform, Pulumi) declares your cloud in version-controlled files, so infra is reviewable, repeatable, and diffable. Learn state, plan/apply, modules, and drift — infra becomes just more code.
Declarative infra (Terraform)
State & drift
plan / apply
Modules & reuse
Build itProvision a VPC + database with Terraform. Someone changes it in the console — how does the next plan reveal and reconcile the drift?
✓ CheckpointExplain what Terraform state is protecting, and what happens when two engineers apply at the same time.
3
Configuration Management
T3 · Advanced · 55 min
Once machines exist, they must be configured consistently. Tools like Ansible push a desired state — packages, files, services — declaratively and idempotently across fleets. Learn playbooks, inventory, and why idempotence is the whole game.
Declarative config (Ansible)
Idempotence
Inventory & roles
Immutable vs mutable infra
Build itWrite a playbook that installs and configures a service on 50 hosts. Run it twice — nothing should change the second time. Why?
✓ CheckpointExplain the difference between configuring a server and replacing it, and why immutable infrastructure sidesteps drift.
4
Kubernetes
T4 · Advanced · 90 min
Kubernetes orchestrates containers at scale by chasing a declared desired state — scheduling, self-healing, rolling out, and load-balancing pods across a cluster. It is the central skill of modern DevOps. Learn pods, deployments, services, and the reconciliation loop.
Pods & deployments
Services & networking
Desired-state control loop
Scaling & rollouts
Build itDeclare 3 replicas behind a service. Kill a pod and push a new image — show how the control loop keeps you available through both.
✓ CheckpointExplain what the control loop does when you delete a pod by hand, and why that is the whole idea.
Pipelines and clusters hold the keys to everything. Learn secret management (Vault, cloud KMS), least-privilege IAM, image and dependency scanning, and supply-chain basics. DevSecOps means security is a step in the pipeline, not a review at the end.
Secret management
Least-privilege IAM
Image & dependency scanning
Supply-chain security
Build itA database password is hard-coded in a repo. Move it to a secret store and inject it at deploy time — with no human ever seeing it.
✓ CheckpointExplain why a secret in an environment variable is better than one in the image and still not good enough.
6
Artifacts & Registries
T6 · Intermediate · 45 min
Between build and deploy live artifacts — container images, packages, binaries — stored in registries with versions and provenance. Learn tagging, immutability, promotion across environments, and why "deploy the exact bytes you tested" matters.
Image & package registries
Semantic tagging
Immutability & promotion
Provenance
Build itThe image that passed staging tests must be the one that reaches prod. How do you guarantee it is byte-identical, not a rebuild?
✓ CheckpointExplain why promoting the same immutable artifact beats rebuilding per environment.
Track 3 of 3
ProductionOperate & scale
6 rungs · ~6h · ends at Cost & Capacity.
0/6 done
1
Observability
P1 · Advanced · 65 min
You cannot operate what you cannot see. Instrument the three pillars — metrics (is something wrong?), logs (what happened?), traces (where?) — with dashboards and the golden signals. Observability is what turns a 3am page into a guided fix.
Metrics / logs / traces
Dashboards
Golden signals
Structured logging
Build itLatency spiked. Show which metric alerts you, which trace localizes the slow hop, and which log lines explain the root cause.
✓ CheckpointExplain which of metrics, logs and traces answers “is it broken”, “why”, and “where” — and why you need all three.
GitOps makes the git repo the single source of truth: a controller continuously reconciles the cluster to match it, so deploys are pull requests and rollbacks are reverts. Pair it with progressive delivery — canary and blue-green — to release safely.
GitOps reconciliation
Canary & blue-green
Automated rollback
Feature flags
Build itRoll out a risky change to 5% of traffic, watch error rate, and auto-roll-back on breach — all driven by a commit to git.
✓ CheckpointExplain what GitOps reconciliation does when someone changes the cluster by hand.
Production breaks; the discipline is how you respond. Learn on-call rotations, alerting that pages on symptoms not noise, incident command, and blameless postmortems that turn outages into permanent fixes. Reliability is a practice, not a state.
On-call rotations
Alerting on symptoms
Incident command
Blameless postmortems
Build itWrite the runbook for a "checkout down" alert: who is paged, the first three checks, and how the postmortem prevents a repeat.
✓ CheckpointExplain why alerting on symptoms beats alerting on causes, and what a page at 3am should mean.
4
SLOs & SRE
P4 · Advanced · 60 min
Site Reliability Engineering makes reliability measurable. Define SLIs, set SLOs, and spend the resulting error budget to balance new features against stability. When the budget is gone, you stop shipping risk and fix reliability — data, not opinions, decides.
SLIs / SLOs
Error budgets
Toil reduction
Reliability vs velocity
Build itSet a 99.9% SLO for an API and derive its monthly error budget in minutes. When the budget is spent, what must the team change?
✓ CheckpointExplain what an error budget lets you say no to — and what it lets you say yes to.
5
Networking at Scale
P5 · Advanced · 60 min
Production traffic needs steering. Learn load balancers (L4/L7), DNS-based routing, CDNs at the edge, TLS termination, and service mesh basics. This is where "it works on one box" becomes "it works for millions across regions."
Load balancing (L4/L7)
DNS routing
CDN & edge
Service mesh basics
Build itServe a global app: put a CDN at the edge, load-balance regional backends, and route users to the nearest healthy region.
✓ CheckpointExplain the difference between L4 and L7 load balancing in terms of what the balancer can see.
The cloud bill is a feature you own. Learn capacity planning, autoscaling, right-sizing, spot vs reserved, and FinOps basics — spotting the idle resources and over-provisioned clusters quietly burning money. Efficiency is reliability you can afford.
Capacity planning
Autoscaling & right-sizing
Spot vs reserved
FinOps & cost visibility
Build itA cluster costs 3x what traffic justifies. Find the waste — idle nodes, over-provisioned requests — and set autoscaling that tracks demand.
✓ CheckpointExplain when spot instances are appropriate and what your workload must tolerate to use them.
DevOps roadmap — frequently asked questions
The common questions before you start — how long it takes, whether to follow it in order, and how it stays current.
How long does this roadmap take?
It runs 18 stations across 3 tracks — roughly ~16h of focused learning, plus the time you spend actually building. It is self-paced, so most people work through it over a few weeks, an evening or a single station at a time.
Do I have to follow the stations in order?
The tracks are ordered so each station builds on the one before, and following them start to finish is the intended path. But every station also stands alone — if you already have the foundations, jump straight to the part you need.
Is it free?
Yes. The whole roadmap, and every handbook, lab, and challenge it links to, is free and open — no sign-up and no paywall.
How is the roadmap kept current?
It teaches the durable fundamentals of the role first, then the tooling and the AI-era shifts on top — so most of it stays relevant as individual tools churn, and it is revised as the role itself changes.
Who is this roadmap for?
Anyone stepping into or leveling up in the DevOps role — whether you are switching in, early-career, or a senior filling gaps. Start where you are; the ladder shows what is left.
Finished this one? 0 / 31 Roadmaps done
Explore the topic
See this alongside everything else on the same subject — handbooks, system designs, challenges and tools, in one place.