Judgement
Frontend roadmap
Not a list of everything that exists — roadmap.sh already is one. This is what I would put on it, in the order I would put it, and why.
What is moving now →
Each card's border says how much it matters:essentialworth itcan wait
98 things across 12 stages
01The ground floor
Nothing here is optional and none of it goes out of date. Most gaps further up the list turn out to be a gap here.
02CSS as a craft
The part people skip because they think they already know it. It is where most of the visible quality lives.
03Accessibility
Its own stage, not a checklist at the end. Retrofitting it costs several times what building with it costs.
04JavaScript in production
The step from making it work to making it survive other people and other months.
05Frameworks
One properly beats five by name. The second one is much cheaper than the first, so the first is the one that matters.
Vue and its ecosystem
Rendering
Other roads
06Tooling
The things around the code. Learn them once, configure them once, and stop thinking about them.
Bundlers
Package managers
Deployment
Linting and formatting
Testing
Repo hosting
Runtime
Documenting
07Beyond the browser
The doors next to yours. None of this teaches you the web, and none of it is where to start — but knowing what exists is worth something.
08Performance
Measured on a real phone on real data, not on your laptop over fibre.
09The browser and the network
Where your code actually runs and how it gets there. Skipping this is why some bugs feel like magic.
10Being useful to the product
The difference between someone who closes tickets and someone people want on the project.
11AI-assisted coding
Already part of the job. It writes faster than you and is confidently wrong, so the skill is not prompting — it is reviewing.
12Judgement
The part no roadmap lists, and the part that separates people after the first two years.