NurseX

Generate unlimited AI quizzes, practice oral exams with Gemini Live, and get detailed insights into your performance. All powered by the Vercel AI Gateway.

Comment

Inspiration

Nursing school is brutal. Beyond memorizing thousands of facts, students face two high-pressure exam formats: written MCQs and the oral viva, where an examiner fires questions at you in real time. Most prep tools only cover static question banks. We wanted something that could generate unlimited fresh questions on any topic, simulate the stress of a live viva, and then tell each student exactly where they're weak — the way a great tutor would. That became NurseX.

What it does

NurseX is an AI-powered nursing exam prep platform with three core features:

  • Quiz Generation — Pick any of 12+ nursing topics, a difficulty level, and a question count. The AI instantly generates fresh MCQs with detailed explanations for every answer.
  • Viva Practice — A simulated oral exam. The AI acts as an examiner, asks follow-up questions, evaluates each spoken/typed answer, and scores your response with constructive feedback.
  • Performance Analytics — A dashboard that tracks every session, breaks down accuracy by topic, and surfaces personalized strengths, weaknesses, and tips for improvement.

It ships with a polished, fully responsive UI and a proper light/dark theme.

How we built it

  • Frontend: Next.js 16 (App Router) + React 19, Tailwind CSS v4, and a custom theme system with persistent light/dark mode.
  • AI: The Vercel AI Gateway via the AI SDK, calling google/gemini-2.5-flash. Using the Gateway means no provider keys are hardcoded — model access is managed centrally and is easy to swap.
  • Persistence: Amazon DynamoDB for quiz/viva history and progress, with anonymous localStorage fallback so users can start instantly with zero signup.
  • Reliability: All AI calls run server-side in Route Handlers with strict input validation and graceful, actionable error states.

For the analytics, accuracy per topic is computed as:

$$\text{Accuracy}_{topic} = \frac{\text{correct answers}}{\text{total questions}} \times 100\%$$

and an overall mastery score weights recent sessions more heavily:

$$M = \frac{\sum_{i

Built With

  • amazon-dynamodb
  • google-gemini-2.5-flash
  • next.js-16
  • react-19
  • recharts
  • tailwind-css-v4
  • typescript
  • vercel
  • vercel-ai-gateway
  • vercel-ai-sdk
  • zod
Share this project:

Updates