Your path is not a course to follow end to end: it is computed. The engine starts from your goal, gathers the skills it requires and their prerequisites, removes what you already master, and offers a next step. With every lesson read, quiz passed or lab validated, it recomputes. This lesson describes those steps, and what changes when you modify a choice.
The five steps of the computation
goal → expected skills → prerequisites → what you know → what is missing → next step- The goal: a job, one or more skills, a certification or a followed course.
- The expected skills: those the goal requires directly, for example a job's core.
- The prerequisites: the engine adds every skill the previous ones depend on, step by step.
- What you know: acquired or declared skills leave the path.
- The next step: among what remains, a skill whose prerequisites are met, with the action that moves it forward.
What is the next step?
The next step is never a locked skill. The engine favours what needs consolidating (a failed quiz), then what you have started, then what awaits its quiz or lab, then an available skill. The action offered is concrete: read a specific lesson, take a quiz, do a lab. It comes with its reason, for example “it unlocks two skills of your path”.
The My path tab shows the result of the computation, here for a platform engineer on the way: the four steps (read, known, done, proof passed), the job's core by domain, then the missions, with the skill to work on to move forward:
Why a course is not imposed in full
A skill is often taught by a few lessons of a course. When your goal keeps only part of a course, its programme and its sidebar switch to the “For my goal” view: only the useful lessons count there, and the Whole path stays one click away. A module can therefore be only partly useful.
What happens if you change tool?
A tool choice (GitHub or GitLab CI, Argo CD or Flux CD, Docker or Podman), set in the “My tools” card of the My goals tab, changes the lessons offered for the skills concerned, and sometimes their prerequisites: installing Kubernetes with kubeadm requires knowing how to manage systemd, with Talos it does not. What you have acquired stays; depending on the skill, it counts for the other tool or not.
What happens if you change job?
Changing job changes the expected core, not your progress: skills already acquired stay acquired, and those both jobs share leave the new path at once. A Linux administrator aiming for site reliability engineering thus finds a good part of the new core already covered.
What the engine does not do
The engine guesses nothing: it only counts what it can measure, that is lessons marked as read, quizzes, placement tests, imported labs and what you declare. It does not grade you or compare anyone: the whole computation happens in your browser.
The computation stays local even with an optional account: the account only copies your progress from one device to another, and it is still the browser that draws the next step from it.
Key takeaways
- The path is recomputed with every lesson, quiz or lab.
- Prerequisites are added automatically; what you know is removed.
- The next step is always playable, with its reason.
- Changing tool or job loses nothing.