RESUMai — AI Resume Analyzer

RESUMai is a Next.js app that analyzes PDF resumes with Google Gemini. Users upload a resume, the backend extracts text, sends it to Gemini, and the UI renders a score plus targeted improvements.

Overview

• Input: PDF resume (max 10MB)
• Processing: PDF text extraction → Gemini analysis → JSON parsing
• Output: Score (0–10), improved bullet points, missing skills, ATS tips, and general suggestions

User Flow

1. Upload a PDF on the landing page.
2. /api/analyze validates file type/size and extracts text (trimmed to ~8000 characters).
3. Gemini returns structured JSON.
4. Results render with copy-to-clipboard actions.

Features

• Drag-and-drop PDF upload with validation
• Fast analysis with structured results
• Copy buttons per section + “copy all”
• Dark, responsive UI with smooth states

Tech Stack

Next.js 14 (App Router), React, TypeScript, Tailwind CSS, Google Gemini API, pdf-parse.

Project Structure

resume-anal/
├── app/
│   ├── api/analyze/route.ts    # Upload + validation + analysis
│   ├── page.tsx                # Main UI
│   └── globals.css             # Global styles
├── components/
│   ├── UploadCard.tsx          # File upload UI
│   └── ResultsCard.tsx         # Results display
├── lib/
│   ├── pdf.ts                  # PDF parsing
│   └── gemini.ts               # Gemini integration
├── public/
├── package.json
└── README.md

Environment Variables

Name	Required	Description
GEMINI_API_KEY	Yes	Gemini API key
GEMINI_MODEL	No	Override model (e.g., gemini-3-flash)

Setup

1. Install dependencies:

   npm install

2. Create .env.local (gitignored):

   GEMINI_API_KEY=your_key_here
   GEMINI_MODEL=gemini-3-flash

3. Start the dev server:

   npm run dev

4. Open http://localhost:3000

API Endpoint

POST /api/analyze
Body: multipart/form-data with field file

Validations

• PDF only (.pdf or application/pdf)
• Max size: 10MB

Success Response

{
  "success": true,
  "data": {
    "score": 8,
    "improvedBullets": ["..."],
    "missingSkills": ["..."],
    "atsTips": ["..."],
    "suggestions": ["..."]
  }
}

Error Responses

• 400: invalid file type/size or empty text
• 500: Gemini or parsing errors

Build & Deploy

npm run build
npm start

Security & Privacy

• Uploaded resumes are processed in memory and not stored.

Troubleshooting

• "GEMINI_API_KEY not configured" → add the key to .env.local and restart.
• "Failed to parse PDF" → use a text-based PDF (not scanned images).
• "File too large" → keep the PDF under 10MB.
• "Only PDF files are supported" → upload a .pdf.