Back openDesk Edu for a sovereign, open-source education — every vote counts.
Vote nowSave products you love by clicking the heart icon.
Every year, people ask the same question: "I want to work with infrastructure, cloud, and DevOps - where do I even start?"
The honest answer hasn't changed: Linux is the foundation. Everything else - containers, Kubernetes, CI/CD, observability, AI infrastructure - runs on Linux. You don't need to be a kernel hacker, but you do need to be comfortable with the command line, understand how processes work, know what a filesystem is, and debug a failing network connection.
This guide maps a complete learning path from zero experience to production-ready DevOps professional - using open-source tools, vendor-neutral certifications, and real-world infrastructure stacks. For more on why fundamentals matter, see Linux, DevOps, and AI: Why the Fundamentals Matter More Than Ever in 2026. For production deployment patterns, see The Open-Source Infrastructure Stack: PostgreSQL, Redis, MinIO, n8n, and Observability in Production. For certification strategy, see DevOps Certifications in 2026: Which Ones Actually Matter for Your Career?.
Before diving in, here's what the industry looks like right now:
| Layer | Dominant Tools | Why It Matters |
|---|---|---|
| Operating System | Linux (Ubuntu, Debian, RHEL, Alpine) | 85%+ of servers, 100% of containers, 100% of cloud |
| Containers | Docker, Podman | Standard unit of deployment everywhere |
| Orchestration | Kubernetes (k3s, K8s, OpenShift) | The cloud OS - runs everything |
| CI/CD | GitHub Actions, GitLab CI, ArgoCD | How code becomes running software |
| Observability | Prometheus, Grafana, Loki | How you know it's actually working |
| Storage | PostgreSQL, Redis, MinIO | The data layer - persistent and cache |
| Automation | n8n, Ansible, Terraform | How you stop doing things manually |
| AI Inference | Ollama, vLLM, Open WebUI | The new workload that needs infrastructure |
Every single one of these runs on Linux. Every single one can be self-hosted. And every single one has a learning path that starts at the command line.
Start here. Do not skip this. Cloud certifications without Linux fundamentals produce engineers who can deploy but cannot troubleshoot.
grep, sed, awk), pipes and redirectsps, top, kill, systemd, journalctlext4, XFS, permissions, linksapt, dpkg, yum, rpmss, curl, pingsudo, file permissions, SSH keys| Resource | Why | Cost |
|---|---|---|
| LPIC-1 Study Guide | Covers exactly what you need - vendor-neutral, practical | Free |
| LPIC-1 101-500 Deep Dive | Command-level walkthroughs for the first exam | Free |
| LPIC-1 102-500 Deep Dive | Shell scripting, networking, security in depth | Free |
| LPIC-1 Practice Exams | Validate knowledge before cert exam | €9.99/month |
| The Linux Command Line (William Shotts) | Best free book for command line fundamentals | Free |
The LPIC-1 certification validates exactly these skills. It's vendor-neutral, internationally recognized, and never expires. Two exams (101-500 and 102-500), 60 questions each, 90 minutes each.
Why LPIC-1 and not RHCSA or CompTIA Linux+? LPIC-1 gives you the broadest foundation. It's distribution-agnostic, costs less, and stacks into LPIC-2, LPIC-3, and the DevOps Tools Engineer certification. You can always add RHCSA later if you move into a Red Hat shop.
Containers are how software is packaged and deployed in 2026. Kubernetes is how it's orchestrated. This phase builds directly on your Linux foundation - containers are Linux processes with namespaces and cgroups.
Dockerfile, multi-stage buildsDo not jump straight into Kubernetes without understanding Docker Compose. k3s is an excellent learning cluster - it runs on a single VM with 2GB RAM and gives you a real Kubernetes API server. Start with Compose for local development, then graduate to k3s for production-like deployments.
Build this as a learning project:
┌──────────────────┐
│ Traefik/Nginx │ SSL termination, routing
│ Reverse Proxy │
└───────┬──────────┘
│
┌──────────────┼──────────────┐
│ │ │
┌──────▼─────┐ ┌─────▼──────┐ ┌─────▼──────┐
│ PostgreSQL│ │ Redis │ │ n8n │ Workflow automation
│ Database │ │ Cache │ │ Workflow │
└────────────┘ └────────────┘ └────────────┘
│
┌──────▼─────┐
│ MinIO │ Object storage (S3-compatible)
│ Storage │
└────────────┘
Every component here runs in a container, connects via Docker networking, and stores persistent data on mounted volumes. Setting this up teaches you more about production infrastructure than any tutorial.
Our production stacks give you exactly these configurations - pre-built, tested, documented:
| Stack | What It Does | Price |
|---|---|---|
| Database Foundation Stack | PostgreSQL + Redis with daily backups | €9.99 |
| n8n Production Stack | Workflow engine with Postgres + Redis | €9.99 |
| MinIO S3 Backup Stack | S3-compatible storage with lifecycle | €9.99 |
| SSL Reverse Proxy Stack | Nginx + Certbot + CrowdSec WAF | €9.99 |
| Observability Stack | Grafana + Prometheus + Loki + Alloy | €9.99 |
Each is a docker compose up -d away from running on your server. You can build them yourself - and you should, for learning - but when you need them in production, having the battle-tested config saves days of debugging.
The LPI DevOps Tools Engineer certification (701-100) covers exactly this territory: containers (Docker, Podman), CI/CD pipelines, infrastructure as code, and monitoring. It's the bridge between traditional Linux administration and modern DevOps practices.
No prerequisites - though practical Linux experience makes it much easier. One exam, 60 questions, 90 minutes.
You've deployed services. Now you need to know they're working, find problems when they break, and measure what matters.
The Observability Stack combines Grafana, Prometheus, Loki, and Alloy into a single Docker Compose deployment. It auto-discovers containers, collects metrics and logs, and comes with pre-built dashboards for node and container monitoring.
This is the same stack used in production environments worldwide - self-hosted, no vendor lock-in, no per-metric pricing.
Observability is not about tools. It's about questions you can answer. Can you answer "what was the error rate 10 minutes before the incident?" without adding instrumentation retroactively? That's what observability buys you.
Manual deployments are error-prone. CI/CD pipelines make them repeatable, auditable, and fast.
n8n connects everything that doesn't have a native integration. It's the glue layer - webhook triggers, Slack notifications, database updates, email parsing, API orchestration. The n8n Production Stack runs it with PostgreSQL persistence and Redis-backed queue processing for reliability at scale.
This is where the industry is heading. AI inference - running LLMs locally - is a new infrastructure workload that combines everything you've learned: Linux, containers, GPUs, networking, storage, and observability.
The typical self-hosted AI stack in 2026 looks like this:
All containerized, all open-source, all running on the Linux infrastructure you've been building since Phase 1.
Here's what a complete production infrastructure looks like when all phases come together:
Every layer here is open-source, self-hostable, and vendor-neutral. Every layer builds on the one below it. And every layer has a learning path that starts with a single command.
| Phase | Certification | What It Validates |
|---|---|---|
| 1 | LPIC-1 | Linux fundamentals, command line, system administration |
| 2 | DevOps Tools Engineer | Containers, CI/CD, IaC, monitoring |
| 3 | LPIC-2 | Advanced Linux, networking, storage, security |
| 4 | LPIC-3 | Specialization: security, HA, virtualization, mixed environments |
You don't need all of them. But each one validates a layer of the stack and makes you more credible when things break.
| Resource | What You Get | Price |
|---|---|---|
| LPI Practice Subscription | Unlimited practice exams, all tracks | €9.99/month |
| DevOps Cheatsheets Bundle | 23 cheatsheets - Docker, K8s, Terraform, more | PWYW from €2.99 |
| Production Stacks | Battle-tested Docker Compose configurations | €9.99 each |
You can build a complete career in DevOps starting from zero - without spending thousands on bootcamps or getting locked into a single cloud vendor. The path is straightforward:
Every step builds on the last. Every step uses open-source tools. And every step gives you transferable skills that work at any company, on any cloud, with any stack.
Start with one command. ls, cd, cat. That's how every DevOps engineer began.
Want this complete learning path as a printable PDF? Download the DevOps Career Roadmap with:
Download the free DevOps Career Roadmap PDF →
| Phase | Stack | What It Gives You | Price |
|---|---|---|---|
| Phase 2: Containers & Data | Database Foundation Stack | PostgreSQL + Redis with automated backups | €9.99 |
| Phase 2: Containers & Data | MinIO S3 Backup Stack | S3-compatible storage with lifecycle policies | Free |
| Phase 2: Containers & Data | SSL Reverse Proxy Stack | Nginx + Certbot + CrowdSec WAF | Free |
| Phase 3: Observability | Observability Stack | Prometheus + Grafana + Loki + Alloy | Free |
| Phase 4: Automation | n8n Production Stack | Workflow automation with Postgres + Redis | Free |
| Phase 5: Certifications | LPI Practice Subscription | Unlimited practice exams for all LPI certs | €9.99/month |
Found this career path helpful? Share it with others who are starting their DevOps journey:
Share on Twitter Share on LinkedIn
This guide was written by Tobias Weiss, a DevOps engineer with over 15 years of experience in Linux, containers, and cloud infrastructure. Tobias creates production-ready Docker Compose stacks and practice exams to help engineers level up their careers.
Need 1:1 career advice? Book a DevOps Career Strategy Session →
Start with free LPIC-1 practice questions →