Build a learning roadmap for a new skill
Your level and weekly hours become staged milestones, each with a build-it project and a test for when you're ready to move on.
Last updated
Fill in the blanks
Assembled in your browser — nothing you type is stored or sent anywhere.
Your prompt
Model-agnostic — works in Claude, ChatGPT, Gemini or any AI chat.
Why this prompt works
- →Requiring a project instead of a tutorial per stage forces active use of the skill, which is what actually produces the competence a roadmap is supposed to build.
- →A checkable "ready for the next stage" test replaces vague progress markers with something you can self-assess against instead of guessing when to move on.
- →Checking my level against the skill's prerequisites first catches a roadmap built on a mismatch — a common reason learning plans stall in week two.
Part by part
- as many as are genuinely needed, don't pad or compress to hit a round number
- Prevents the five-stage roadmap that exists because five is a satisfying number rather than because the skill has five stages.
- one project that forces you to use it (not a tutorial to follow)
- The distinction that decides whether the roadmap works. Following a tutorial produces the feeling of learning; building something produces the ability.
- roughly how long it takes at my pace
- Ties estimates to the stated hours rather than to a full-time schedule, which is what most published roadmaps quietly assume.
- you're ready for the next stage when...
- A self-administered test. Without one, people move on when a stage becomes boring rather than when it is finished.
- say so before building a roadmap around that gap
- Catches the mismatch between stated level and actual prerequisites, which is the usual reason a roadmap stalls at stage two.
Common mistakes
Give me a roadmap to learn machine learning.With no level and no time budget, the output is a generic curriculum assuming full-time study. It is discouraging rather than wrong, and it is abandoned in week two.
What should I learn to become a data scientist?Produces a list of topics with no sequence, no projects and no way to tell whether you have learned any of it. A list is not a roadmap, and the ordering is most of the value.
Build me a 30-day learning plan.Fixing the duration first forces the content to fit the calendar. Some skills do not compress, and a plan that pretends otherwise produces a month of surface familiarity.
Questions
Why projects rather than tutorials?
Because a tutorial supplies the decisions. Following one produces working code and very little transferable skill, since the hard part — deciding what to do when something does not work — was removed.
Are the time estimates reliable?
Directionally at best. They are useful for comparing stages against each other rather than for planning a calendar, and most people take longer than any estimate suggests.
Why does each stage need a readiness test?
Because without one, the decision to move on is made on boredom or confidence, neither of which correlates with competence. A concrete check is the difference between a roadmap and a reading list.
What if my stated level is wrong?
The prompt asks it to say so, which is the point. Overstating your level produces a roadmap that starts above you and fails at the first stage, and that failure reads as being bad at the skill rather than at estimating.