| Path | Page | Description |
|---|---|---|
/ |
SignInPage | Candidate sign-in |
/dashboard |
DashboardPage | Candidate dashboard |
/exam-test/1 |
ExamTestPage | Exam session — step 1 |
/exam-test/2 |
ExamTestPage | Exam session — step 2 |
/exam-test/3 |
ExamTestPage | Exam session — step 3 |
/complete |
CompletePage | Exam completed confirmation |
/timeout |
TimeoutPage | Exam timed out |
undefinedHave you worked with any MCP (Model Context Protocol)?undefined
Yes, I used the Figma MCP (claude.ai Figma) directly inside Claude Code. It allowed me to pull design context, screenshots, and component metadata straight from a Figma file without leaving the terminal. The workflow was: share a Figma URL → get_design_context returns the node’s structure, a screenshot, and code hints → adapt the output to the project’s stack (React + Tailwind). This eliminated the back-and-forth of manually inspecting frames and copying values.
undefinedWhich AI tools or processes have you used or recommend to speed up frontend development?undefined
I prefer Claud Code.
If a candidate loses internet during an exam, the strategy is:
undefinedPersist state to localStorage continuously — on every answer change and every timer tick, write the current exam state (answers, timeLeft, currentQuestion) to localStorage keyed by exam ID and candidate session. Zustand’s persist middleware makes this a one-liner.
undefinedDetect connectivity with navigator.onLine + online/offline events — show a non-blocking banner (“You are offline — your progress is saved locally”) so the candidate is aware but not panicked.
undefinedQueue answer submissions — instead of firing API calls immediately, push submissions into an in-memory/localStorage queue. A background sync loop (or the Background Sync Service Worker API) drains the queue once connectivity is restored.
pnpm i
pnpm dev
We use cookies
We use cookies to analyze traffic and improve your experience. You can accept or reject analytics cookies.