Skip to main content

​
Welcome to Resume Generator

Resume Generator is an intelligent interview preparation platform that leverages Google’s Gemini AI to help job seekers ace their interviews. Upload your resume and target job description to receive a comprehensive interview strategy tailored specifically for you.

Smart analysis

AI-powered matching that compares your profile against job requirements and calculates a compatibility score

Interview questions

Get both technical and behavioral questions you’re likely to face, with detailed answer strategies

Skill gap analysis

Identify areas where your profile falls short and understand the severity of each gap

Preparation roadmap

Receive a day-by-day preparation plan with specific tasks to maximize your interview readiness

​
How it works

The platform combines resume parsing, natural language processing, and AI-driven analysis to create actionable interview preparation reports.
1

Upload your information

Provide a job description along with either your resume (PDF/DOCX) or a self-description of your experience
2

AI analysis

Google Gemini analyzes the job requirements against your profile using structured data extraction with Zod schemas
3

Receive your report

Get a comprehensive interview report including match score, predicted questions, skill gaps, and a preparation timeline
4

Export tailored resume

Generate an ATS-friendly resume PDF optimized for the specific job using Puppeteer

​
Key features

​
Multi-format resume parsing

The backend uses pdf-parse for PDF files and mammoth for DOCX files to extract text. If local parsing fails (e.g., scanned PDFs), the system falls back to sending the raw file directly to Gemini’s multimodal API. 
const isPdfMagic = b.length >= 4 && b[0] === 0x25 && b[1] === 0x50 && b[2] === 0x44 && b[3] === 0x46
const isZipMagic = b.length >= 4 && b[0] === 0x50 && b[1] === 0x4b && b[2] === 0x03 && b[3] === 0x04
Source: backend/src/controllers/interview.controller.js:45-47

​
Structured AI responses

The application uses Zod schemas with zod-to-json-schema to ensure AI responses conform to expected data structures: 
const interviewReportSchema = z.object({
  matchScore: z.number().describe("A score between 0 and 100"),
  technicalQuestions: z.array(z.object({
    question: z.string(),
    intention: z.string(),
    answer: z.string()
  })),
  behavioralQuestions: z.array(z.object({
    question: z.string(),
    intention: z.string(),
    answer: z.string()
  })),
  skillGaps: z.array(z.object({
    skill: z.string(),
    severity: z.enum(["low", "medium", "high"])
  })),
  preparationPlan: z.array(z.object({
    day: z.number(),
    focus: z.string(),
    tasks: z.array(z.string())
  }))
})
Source: backend/src/services/ai.service.js:11-33

​
Resume PDF generation

Generate professionally formatted, ATS-friendly resume PDFs tailored to specific job descriptions using Puppeteer for HTML-to-PDF conversion. 
const browser = await puppeteer.launch()
const page = await browser.newPage()
await page.setContent(htmlContent, { waitUntil: "networkidle0" })

const pdfBuffer = await page.pdf({
  format: "A4",
  margin: {
    top: "20mm",
    bottom: "20mm",
    left: "15mm",
    right: "15mm"
  }
})
Source: backend/src/services/ai.service.js:79-92

​
Tech stack

Frontend

  • React 19.2 with Vite for fast development
  • React Router 7 for client-side routing
  • Axios for API communication
  • Sass for styling

Backend

  • Node.js with Express 5
  • MongoDB with Mongoose ODM
  • JWT authentication with httpOnly cookies
  • Google Gemini AI via @google/genai

File processing

  • Multer for multipart/form-data uploads
  • pdf-parse for PDF text extraction
  • mammoth for DOCX parsing
  • Puppeteer for PDF generation

Data validation

  • Zod for runtime type validation
  • zod-to-json-schema for AI response schemas
  • bcryptjs for password hashing

​
Architecture

The application follows a modern full-stack architecture:
  • Frontend: React SPA running on http://localhost:5173 with context-based state management
  • Backend: RESTful API on http://localhost:3000 with JWT-based authentication
  • Database: MongoDB for storing user accounts and interview reports
  • AI Integration: Google Gemini (gemini-3-flash-preview model) for intelligent content generation
The frontend and backend run as separate processes. CORS is configured to allow credentials from the Vite dev server.

​
What’s next?

Ready to get started? Follow our quickstart guide to generate your first interview report, or check out the detailed installation instructions to set up the development environment.

Build docs developers (and LLMs) love

Get started for freeTalk to us