~/codecrush/maps
Career maps, with the market alongside.
Study trails that do not stop at a list of technologies: each topic has a short lesson, a mini-test and a check that has to be earned — plus the shortcuts to the glossary entry, to the open jobs in that area and to the tool that lets you practise the concept.
Available trails
## frontend
Career map: Frontend
Frontend is the most sought-after entry point in development, and therefore the most crowded. The most common mistake is jumping straight to the framework: React solves a lot, but technical interviews ask about what sits underneath — how the browser loads a page, how JavaScript handles asynchronicity, why the layout broke.
5 stages · 22 topics · 14 essential · 110 questions
## backend
Career map: Backend
Backend holds the highest salary ranges in application development, and the reason is direct: the cost of a mistake is higher. What breaks there takes down the product, leaks data or loses money — which is why interviews probe data modelling and reasoning about failure far more than framework syntax.
5 stages · 21 topics · 11 essential · 105 questions
## dados
Career map: Data
Data is the area that grew most in roles and salary in recent years, and also the one that produces the most wrong study decisions — because the term covers three different professions. Data engineering builds the path the data travels; analytics and BI turn data into decisions; data science models and predicts. They demand different foundations.
5 stages · 20 topics · 9 essential · 100 questions
Why another roadmap
Career roadmaps exist, and the best ones are in English. Their problem is not the quality of the list — it is that the list ends in itself: you find out you need to learn "observability" and end up exactly where you were, without knowing what the word means in practice, how many jobs actually ask for it, or where to practise.
Here every topic carries three pointers to what already exists on the site: the glossary entry when the term is new, the jobs panorama for that area — with open volume and observed salary range — and the tool that lets you practise the concept in the browser.
How to use it
- Pick a trail and follow the order of the stages. It is not easiest to hardest — it is the order in which each block unlocks the next in a real interview.
- Cover the essentials before the recommended ones. Essentials appear in practically every job ad; recommended is what gets you past screening; optional is specialisation, and picking one beats attempting them all.
- Earn the check, do not tick it. Each topic opens a two-minute lesson and a 5-question test; the topic is only marked complete from 80% correct. Progress is saved in your browser so you can pick up where you left off.
- Apply before you finish. Waiting for the whole map is the most expensive mistake: overstated requirements in job ads are common and often negotiable.
Why there is a test, and why it is grumpy
The previous version of this map had tick boxes. The problem is that a tick box records a willingness to click, and a willingness to click is not knowledge — you could complete the whole trail in thirty seconds and walk away with exactly what you walked in with.
Now each topic has a short lesson, the classic trap of that subject and 5 questions written from mistakes that show up in code review and in interviews. Scoring 80% marks the topic; below that the test tells you that you did not pass, with no hedging and no participation prize. Failing erases nothing: you reread, you retake — and the questions come shuffled, so memorising positions does not help.
The feedback is deliberately blunt because the opposite is worse. Praising four out of ten teaches that four out of ten is fine, and the interview will disagree in a considerably more expensive way.
What these trails are not
They are not a course syllabus and not a promise of employment. They are an opinionated reading of what the Brazilian market is asking for, built from what shows up in the ads CodeCrush aggregates every day. Where the reading is arguable, the text says so — and where there is a number, it comes with its source and sample size.