9 career tracks · 324 validated labs · 15 tool hubs
Learn by
doing it
From the Linux shell to Kubernetes, Python pipelines, RAG and frontend work: every lab is a real task in a sandbox in your browser, graded on the machine, not on multiple choice. A reference hub for each tool, an AI tutor that can run offline, and a certificate at the end of the track.
The catalog
Nine tracks,
one curriculum
DevOps, AI engineering, MLOps, data, backend and frontend — plus the Codex, Claude and Copilot tracks for people who build with AI coding agents. Each track is a sequence of validated labs that ends in a certificate.
PATH 01
DevOps Engineer
PATH 02
AI Engineer
PATH 03
MLOps Engineer
PATH 04
Data Engineer
PATH 05
Backend Engineer
PATH 06
Frontend Engineer
PATH 07
Codex Engineer
PATH 08
Claude Engineer
PATH 09
Copilot Engineer
The method
A real terminal.
A real validator.
A certificate at the end.
Hands-on,
not hands-waving
Every lab is a real command in a disposable sandbox — a shell, a Kubernetes cluster, a Terraform state, a Git repository, a Python environment. You deploy the container, apply the manifest, break the config, and read the actual error.
Then an automated validator inspects the machine: the port is mapped, the pod is Running, the metric is scraped, the file holds the right data. It grades the system, not the answer key. Three of those labs, from three different tracks:
Beyond the labs
The rest of
the platform
The labs are the core. Around them sit the references, the tutor and the record — the parts that keep you moving when a lab fights back.
Tool hubs · 15
A reference for every tool
Docker, Kubernetes, Terraform, AWS, Git, CI/CD, Prometheus, ArgoCD, networking, RAG, agents, prompt engineering, vLLM, MLflow. Each hub carries the architecture, a cheat sheet, the failure modes people actually hit, and the labs that use it.
Open the tool hubsAI chat · 4 sources
A tutor you can run offline
Stuck at midnight? Ask about the lab. Download a model once and it answers in your browser on your own GPU, or point the chat at a local Ollama server, the hosted model, or your own key.
Open the chatProgress · XP & certificates
Work that leaves a trace
Every validated lab earns XP and feeds a streak, a level and an achievement. The dashboard tracks the track you are on; clearing it unlocks a printable certificate.
See the dashboardBy the numbers
What 324 labs
actually looks like
- DevOps Engineer72
- AI Engineer33
- Data Engineer32
- Backend Engineer32
- Frontend Engineer32
- Codex Engineer32
- Claude Engineer32
- Copilot Engineer32
- MLOps Engineer27
9
Career tracks
75
Skills
188
Topics
324
Validated labs
The price
Free forever
No card, no seat licence, no trial that expires. The whole curriculum is MIT licensed and self-hostable, and it stays up as long as the project does.
Start a trackHow it works
Your first week,
step by step
- 01
Pick a track
Nine of them — DevOps, AI engineering, MLOps, data, backend, frontend, and the Codex, Claude and Copilot tracks for AI coding agents.
- 02
Do the first lab
A beginner task of a few minutes, with every command explained before you type it.
- 03
Build the streak
One lab a day beats a weekend binge. The streak, the XP ladder and the achievements make the habit visible.
- 04
Finish with proof
Clear every lab in a track and the certificate unlocks — a printable record of the work you actually did.
Keep going
Pick up where
you left off
The labs are ordered into a curriculum, so the next one is always waiting. One new lab drops every midnight.
Lab of the dayNetworking
Add a Health Check to the Proxy
A proxy is only as good as its health check. Start the application, prove it answers, and script the exact same question so the proxy can keep asking it.
Create a Project Structure
Create a project directory containing a src folder, a tests folder and a README.md. Follow the walkthrough below — every command is explained, and the validator checks what you actually built.
Find the Largest Log
Something is writing huge log files to /var/log/app. Inspect the filesystem, find the largest .log file in that directory and record it in ~/report/largest.txt. The walkthrough shows every command.
Lock Down a Secret File
Create a secrets file with restrictive permissions. Only the owner should be able to read and write it — nobody else. This is exactly how SSH private keys work.
Count Errors in a Log
The application log at /var/log/app/app-2026-09-27.log is full of noise. Use grep and pipes to count the ERROR lines and save the result.
Questions
Frequently
asked
Everything a first-time learner usually wants to know before they start.
Is this only about Linux and DevOps?
No. Linux is one hub among fifteen. The curriculum runs from DevOps and data engineering to AI engineering, MLOps, backend, frontend and three AI coding-agent tracks, and the tool hubs cover Docker, Kubernetes, Terraform, AWS, Git, CI/CD, Prometheus, ArgoCD, networking, RAG, agents, prompt engineering, vLLM and MLflow.
Do I need my own machine, a cloud account or a credit card?
None of them. Every lab runs in a sandbox inside your browser — a real shell with docker, kubectl, terraform, git and Python available. There is nothing to install and nothing to pay for.
I have never opened a terminal. Will I be lost?
No. Every track starts with a beginner lab that takes a few minutes, and each command comes with a one-line explanation of what it does and why.
How do I know whether I actually learned it?
The validator checks the state of the sandbox, not your answers. If a container is not running with the right port mapped, or the pod is not Running, the task is not passed however confident you feel.
Do I need a GPU or an API key for the AI chat?
No. Download a small model once and it answers in your browser on your own GPU, connect a local Ollama server, use the hosted model, or bring your own OpenAI-compatible key. The labs themselves never call a model.
How long before I am ready for a job?
It depends on the track and on what you already know. A focused track is roughly 27 to 72 labs — at one or two labs a day, that is a couple of months of steady work.
Can my team or my class use this?
Yes. Sign in with GitHub to sync progress across machines, or self-host the platform for an onboarding cohort. The curriculum is MIT licensed.
Is it really free, or is there a catch?
There is no catch and no trial countdown. The platform is open source and free to use, including every lab in every track.
Ready when you are
Stop watching.
Start shipping.
Pick a track and open the first lab in a real sandbox. Free, no sign-up, two minutes from now.