For front end engineers to learn & connect. 🌬️Breeze thru your interviews w/ GreatFrontEnd Interviews 🚀Build project challenges w/ GreatFrontEnd Projects
Joined September 2022
- Tweets1.2K
- Following316
- Followers9.5K
- Likes75
5 ways to show senior level impact on a front end resume.
Start with the formula. “Led front end development” names a role. Owned [scope] under [constraint], chose [direction], produced [result] names a scope, a risk and a decision.
“Built design system” becomes the APIs you owned plus who adopted them. A component count shows what you built; adoption shows how widely it’s used.
And turn “mentored developers” into an artifact the team still uses, like a review checklist.
5 async patterns that sound similar and behave differently:
Promise.all vs Promise.allSettled
Promise.race vs Promise.any
async await vs .then
setTimeout vs setInterval
a microtask vs a macrotask
Follow for more.
#javascript #async #promises #webdevelopment #frontend
Snowflake asked a front end candidate to build a grid game, reported in the GreatFrontEnd community.
Model the board in state first: robot {row, col}, candy {row, col}, wins. The grid is data, the DOM just draws it.
Arrow keys move the robot, so listen for keydown.
The robot can’t leave the grid, so clamp every move. Walking into a wall is a no-op, not an error.
When the robot reaches the candy: store the score and the positions, derive the catch. Compare after each move, and if they match, increment and respawn.
Your CLAUDE.md is quietly running on instructions you wrote months ago.
A path moves, a build command changes, you upgrade the model, a skill gets added. Nobody goes back to fix the line, so the stale one stays and steers every session.
Claude Code can read your instruction files and flag what’s stale. The normal /doctor checks your install; add prompt-audit and it reads your instructions instead. Needs v2.1.283 or later.
It surfaces outdated lines and conflicting ones across files.
5 JavaScript cancellation patterns that look similar:
xhr.abort() vs fetch() + AbortController
setTimeout vs AbortSignal.timeout()
manual signal chaining vs AbortSignal.any()
signal.aborted vs the abort event
one generic catch vs error.name
Follow for more.
Node.js fixed a Brotli bug that can let attackers run code on a server.
Five updates for front end engineers from the last two weeks.
01 Node bundles its own copy of Brotli, so it isn’t in package.json and npm audit can’t see it. Fixed in 22.23.4, 24.22.1 and 26.12.1.
03 Fake take-home assignments ship their own Tailwind plugin to run hidden code on your machine. npm install pulls a private package via the repo’s own .npmrc, and the dev server loads it. --ignore-scripts won’t stop it.
Run take-homes in a container.
04 Developers are ranking what Interop 2027 should make work everywhere. Josh Comeau’s top three: @function, corner-shape, border-shape.
05 Chrome 156 lets a failed import() be retried. Before 156 it cached the failure and the retry failed instantly.
Modal vs popover: the difference is what happens to the rest of the page.
A popover leaves it interactive. dialog.showModal() adds a ::backdrop and makes the rest of the document inert, so nothing outside the dialog can be interacted with.
A popover is always non-modal. popover="auto" closes on outside click or Esc, popover="manual" you toggle yourself.
Both rise into the top layer, so neither needs a z-index. The z-index: 999 you were about to write does nothing for either one.
5 DOM event pairs that sound similar:
Individual event listeners vs event delegation
Event capturing vs event bubbling
event.target vs event.currentTarget
preventDefault() vs stopPropagation()
removeEventListener() vs AbortController with signal
Follow for more.