The Honest Reason React Developer Rejections Rarely Get Explained

CANDIDATE SCORECARD (example format)

JavaScript fundamentals   : 4 / 5
React and hooks           : 3 / 5
Problem solving           : 3 / 5
Code quality              : 4 / 5
Communication             : 3 / 5

Notes:
- Clean component split, good naming.
- Fetched data in useEffect with no cleanup;
  did not see the race condition when asked.
- Needed two hints to derive the filtered list
  instead of storing it in state.

Recommendation: Lean no hire for this level.

Dear Candidate,

Thank you for taking the time to interview for the
React Developer position. After careful consideration,
we have decided to move forward with other candidates
whose experience more closely matches our needs.

We wish you the best in your search.

---
Notice: every useful detail from the scorecard,
the effect cleanup gap and the derived state hint,
has disappeared.

function ProductSearch({ products }) {
  const [query, setQuery] = useState("");
  const [filtered, setFiltered] = useState(products);

  // Derived state kept in sync by an effect:
  // an extra render on every keystroke, and a
  // stale list if products change and the deps
  // are wrong.
  useEffect(() => {
    setFiltered(
      products.filter((p) =>
        p.name.toLowerCase().includes(query.toLowerCase())
      )
    );
  }, [products, query]);

  // ...
}

function ProductSearch({ products }) {
  const [query, setQuery] = useState("");

  // Derived during render: always in sync, no effect.
  const q = query.trim().toLowerCase();
  const filtered = q
    ? products.filter((p) => p.name.toLowerCase().includes(q))
    : products;

  // If profiling showed this was expensive for a
  // very large list, wrap it in useMemo with
  // [products, q] as dependencies.

  return (
    <ProductGrid
      query={query}
      onQueryChange={setQuery}
      items={filtered}
    />
  );
}

WRITE WITHIN AN HOUR OF THE INTERVIEW

Planned length: ____  Actual length: ____

Questions asked (as close to exact as possible):
1. _______________________________________
2. _______________________________________
3. _______________________________________

Where did I need hints? _____________________
Which question was asked twice? _____________
Where did they go deeper? ___________________
Where did they move on quickly? _____________

The moment I felt least sure: _______________

At the end of a technical round, it is fine to ask:

'Is there anything in my solution you would have
liked me to approach differently?'

or

'Was there a part of the task where you expected
something I did not cover?'

Many interviewers will answer honestly in the room,
even when the company will not send written feedback
later. Listen without arguing and write it down.

RATE EACH 1-5, BASED ON THE ACTUAL ROUND

JavaScript  : explained closures/async clearly?   __
React model : state, props, effects correct?      __
Effects     : no derived state, cleanup handled?  __
Problem     : solved without many hints?          __
Code        : readable names, small components?   __
Testing     : could discuss how to test it?       __
Communicate : answer-first, clear, concise?       __
Project     : explained decisions and trade-offs? __

Lowest score = most likely reason.

AUDIT EVERY useEffect IN YOUR PROJECT

For each effect ask:
[ ] Does it talk to something outside React?
    (network, timer, subscription, DOM API)
    If not -> can it be derived during render?
[ ] Does it have a cleanup if it starts something?
[ ] Are all dependencies listed honestly?
[ ] Can a slow response overwrite a newer one?
[ ] Would it still be correct if it ran twice?
    (React Strict Mode runs it twice in development)

FROM LOWEST SCORE TO ACTION

Effects -> refactor your project's effects using
  the audit, and explain each change aloud.
Problem solving -> redo the interview task alone
  this week, without looking anything up.
Communication -> record five answers, answer-first.
Testing -> write three tests for your main form.
Project -> write the decisions section of your README.

Book one mock round to check the change worked.