Frontend Developer Mock Interview Guide
How to prepare for frontend interviews beyond framework trivia: browser fundamentals, performance, accessibility, state and explaining your decisions.
HotSeat team
Updated November 2026 3 min read
Frontend interviews jump around. You can go from closures to keyboard navigation to why you'd pick one state library over another in about ten minutes. Framework knowledge won't carry you through that on its own, and frameworks change anyway. What holds up is understanding how the browser works and being able to explain what your code does for the person using it.
Before anything else, refresh the basics: semantic HTML and forms, how the browser loads and renders a page, CSS layout and specificity, and JavaScript scope, async behaviour and error handling. Then practise the questions below, because they're the ones that show how you actually work.
In this article
‘A page feels slow. What do you check first?’
If you open with ‘I'd add memoisation’, you've failed the question. The interviewer wants to hear you measure before you change anything. Start by asking which interaction feels slow and for whom. Is it the first load, or clicking a filter? Those have different causes.
Then open the browser's performance tools and work out whether the time is going on the network, rendering or JavaScript. Lighthouse is a good place to start, but its score is only a clue. Use a metric that matches the complaint, like Largest Contentful Paint for a slow load or Interaction to Next Paint for a sluggish click.
Mention slow devices before they do. Your laptop is probably far faster than what many of your users have, so test with CPU and network throttling or on a low-end phone. And fix the biggest delay people can actually feel, rather than shaving milliseconds where nobody will notice.
‘How do you decide where state should live?’
Keep it as close as you can to where it's used, and only move it up when something else genuinely needs it. Then talk about the cost of going global: more components re-render, it's harder to work out what changed a value, and you may have to clean it up by hand when it's no longer needed.
Expect ‘what would make you change your mind?’ as the follow-up. It helps to have a real example, like form state that had to move up when a second step needed it. Server data is also worth mentioning separately, since it comes with caching and staleness problems that local UI state doesn't have.
‘How would you make a custom menu accessible?’
Start by asking whether it needs to be custom, because a native element gives you keyboard support and screen reader behaviour for free. If it does, walk through the keyboard behaviour, where focus goes when it opens and closes, and how the name and state get announced.
‘Why not just add ARIA to a div?’ is a likely follow-up. ARIA changes what's announced but adds no behaviour, so you'd still have to build all the keyboard handling yourself. Then say how you'd test it: unplug the mouse, and try it with a screen reader. Mentioning ARIA is easy, so talking about how you'd test it will help you stand out.
A project story that does the job
Interviewers often ask for a time you improved something for users. Here's an invented example that covers the problem, the choice and the check:
“Our custom checkout dialog worked fine with a mouse, but keyboard focus could move behind it. I replaced the hand-built focus logic with the native dialog element where it was supported, added a tested fallback, then went through the flow using only a keyboard and a screen reader. It fixed the audit issue and let keyboard users complete checkout without losing their place.”
It starts with who was affected, and the component tree doesn't come up at all. Prepare one story like this, and be ready to explain what you'd have done if the native element wasn't an option.
Testing and working with other people
When they ask about testing, say what a test protects. ‘I test the checkout flow because that's where a bug costs us money’ tells them more than any coverage percentage. It's also fair to say what you wouldn't test.
You'll probably get a question about disagreeing with a designer or dealing with an API that doesn't give you what you need. Have a real example ready, ideally one where you changed your mind or found a middle ground.
Practise connecting browser knowledge to real users
Try a frontend mock interview with technical questions, project follow-ups and no tab open with the answer.
Start a frontend interviewSources and further reading
- Accessibility principles (W3C Web Accessibility Initiative)
- Accessibility (MDN Web Docs)
- Why web performance matters (MDN Web Docs)