CV Modernizer

A privacy-first, AI-powered resume modernization utility. Upload an old PDF/DOCX, let Gemini parse it into a structured editor, pick a template, and download — all in one session. No account required. No files stored.

🚀 Live Demo: freeresume.site

Overview

CV Modernizer is a stateless web utility that lets you:

Modernize an outdated PDF or DOCX resume into a clean, editable structure
Edit every section (bio, experience, education, skills, links) in a live side-by-side editor
Reorder entries within sections using ↑/↓ controls
Preview your resume rendered in multiple professional templates with theme colours
Download the finished resume (PDF via Playwright / DOCX via python-docx)

No login. No cloud storage of your resume. No data sent anywhere beyond what is strictly needed.

Screenshots

🏠 Landing Page

✏️ Live Side-by-Side Editor

Key Features

Feature	Details
📄 File Upload	Accepts PDF and DOCX up to 5 MB
🤖 AI Extraction	Gemini parses raw text → typed JSON (bio, experience, education, skills, links)
✏️ Live Editor	Side-by-side accordion editor with instant preview
↕️ Reordering	Move experience, education, and skill groups up/down
🎨 Templates	5 templates (Minimalist, Modern Tech, Executive, Creative, Elegant)
🖌️ Theme Colours	12 curated colour palettes
🛡️ Rate Limiting	3-hour cooldown per email via Supabase (prevents quota abuse)
📱 Mobile Ready	Fully responsive with floating view-toggle and A4 scaling
🔒 Privacy-first	Full resume content never persisted; only minimal metadata cached
☕ Donations	UPI QR (India) and Ko-fi (international)

Architecture

┌─────────────────────────────────────────────────────┐
│                  Browser (Next.js)                  │
│  ┌──────────────┐        ┌───────────────────────┐  │
│  │   Editor.tsx │◄──────►│  ResumePreview.tsx    │  │
│  │  (sidebar)   │  state │  (live PDF-like view) │  │
│  └──────┬───────┘        └───────────────────────┘  │
│         │ POST /api/extract (FormData)               │
└─────────┼───────────────────────────────────────────┘
          │
          ▼
┌─────────────────────────────────────────────────────┐
│              FastAPI Backend (port 8000)             │
│                                                     │
│  1. Read file bytes (in-memory, never to disk)      │
│  2. Extract text  ← PyMuPDF (PDF) / python-docx     │
│  3. Regex guard   ← reject if no email or phone     │
│  4. Rate-limit    ← Supabase TTL check (3h window)  │
│  5. Gemini call   ← structured JSON extraction      │
│  6. Cache write   ← minimal record to Supabase      │
│  7. Return JSON   ─────────────────────────────►    │
└─────────────────────────────────────────────────────┘
          │                        │
          ▼                        ▼
   ┌────────────┐          ┌──────────────────┐
   │  Supabase  │          │  Gemini API      │
   │  cv_cache  │          │  (1.5 Flash / 8b)│
   └────────────┘          └──────────────────┘

Tech Stack

Frontend

Package	Purpose
Next.js 14 (App Router)	React framework, SSR
TypeScript	Type safety throughout
Vanilla CSS	Custom design system (no Tailwind)
qrcode.react	Live UPI QR code generation

Backend

Package	Purpose
FastAPI	REST API framework
Uvicorn	ASGI server with hot-reload
google-genai	Google Gemini SDK (new v1.x)
PyMuPDF (fitz)	In-memory PDF text extraction
python-docx	In-memory DOCX text extraction
playwright	Server-side headless browser for pristine PDF rendering
supabase-py	Supabase client (rate-limit cache)
python-dotenv	`.env` loading
python-multipart	Multipart file upload parsing

Infrastructure

Service	Purpose
Supabase (Postgres)	`cv_cache` table for rate-limiting
Google AI Studio	Gemini API key

Project Structure

CV Modernizer/
├── backend/
│   ├── main.py              # FastAPI app — full extraction pipeline
│   ├── requirements.txt     # Python dependencies
│   ├── schema.sql           # Supabase table + index definitions
│   └── .env                 # Secrets (not committed)
│
├── frontend/
│   ├── public/
│   │   └── upi-qr.png       # UPI QR fallback image (replaced by live QR)
│   └── src/
│       ├── app/
│       │   ├── page.tsx         # Root page (renders <Editor />)
│       │   ├── layout.tsx       # HTML shell, metadata
│       │   ├── globals.css      # Design system — tokens, layout, components
│       │   └── resume.css       # Resume template print styles
│       ├── components/
│       │   ├── Editor.tsx       # Main editor + upload flow + support section
│       │   └── ResumePreview.tsx# 5 resume templates rendered as styled HTML
│       └── types/
│           └── resume.ts        # Shared TypeScript types + defaultResumeData
│
└── README.md

Getting Started

Prerequisites

Node.js ≥ 18
Python ≥ 3.10
A Gemini API key (free tier works) — get one here
A Supabase project (free tier) — supabase.com

1. One-Click VPS Deploy (Recommended)

The easiest way to deploy is using the automated script:

git clone <your-repo-url>
cd "CV Modernizer"
# For IP-based access:
bash deploy.sh
# For Domain-based access (Caddy/SSL):
bash deploy.sh yourdomain.com

2. Manual Setup (Local Development)

Prerequisites

Node.js ≥ 20 (Required for Next.js 15+)
Python ≥ 3.10
A Gemini API key — get one here
A Supabase project — supabase.com

Backend Setup

python -m venv .venv
.venv\Scripts\activate          # Windows
# source .venv/bin/activate     # macOS/Linux

pip install -r backend/requirements.txt
playwright install chromium
cp backend/.env.example backend/.env
# Edit backend/.env with your keys
cd backend && uvicorn main:app --reload

Frontend Setup

cd frontend
npm install
npm run dev

Open http://localhost:3000 in your browser.

Environment Variables

All secrets live in backend/.env. Never commit this file.

# ── Gemini AI ──────────────────────────────────────────────────────
GEMINI_API_KEY=AIza...

# Model to use for extraction (default: gemini-1.5-flash-8b)
# Options (pick one):
#   gemini-1.5-flash-8b   — lightweight, generous free-tier quota (default)
#   gemini-1.5-flash-002  — full Flash, better at complex CVs
#   gemini-2.0-flash-lite — latest, if your key has access
GEMINI_MODEL=gemini-1.5-flash-8b

# ── Supabase ────────────────────────────────────────────────────────
SUPABASE_URL=https://your-project-id.supabase.co
# IMPORTANT: Use the SERVICE ROLE key, NOT the anon key.
# The service role key bypasses Row Level Security — backend only, never expose to frontend.
SUPABASE_KEY=eyJ...

Warning

Always use the service role key in SUPABASE_KEY, not the anon key. The backend uses it server-side and it is never sent to the browser.

API Reference

Base URL: http://localhost:8000

`GET /api/health`

Liveness probe. Returns which env vars are configured.

{
  "status": "ok",
  "gemini_key_set": true,
  "supabase_url_set": true,
  "supabase_key_set": true
}

`POST /api/extract`

Main extraction endpoint. Accepts a PDF or DOCX file and returns a fully structured resume JSON.

Request: multipart/form-data

Field	Type	Description
`file`	File	PDF or DOCX, max 5 MB

Success Response 200:

{
  "resume_data": {
    "bio": { "name": "...", "title": "...", "email": "...", "phone": "...", "location": "...", "summary": "..." },
    "experience": [
      { "id": "exp1", "role": "...", "company": "...", "location": "...", "startDate": "...", "endDate": "...", "bullets": ["..."] }
    ],
    "education": [
      { "id": "edu1", "degree": "...", "institution": "...", "startDate": "...", "endDate": "..." }
    ],
    "skills": [
      { "category": "Languages", "items": ["Python", "TypeScript"] }
    ],
    "links": [
      { "label": "GitHub", "url": "https://github.com/..." }
    ]
  },
  "detected_email": "user@example.com",
  "detected_phone": "+91 9999999999",
  "raw_text_length": 3420
}

Error Responses:

Status	Meaning
`400`	Invalid file type (not PDF/DOCX)
`413`	File exceeds 5 MB limit
`422`	No email or phone found in document
`429`	Same email processed within the last 3 hours
`503`	Gemini API key missing or quota exceeded
`500`	Unexpected server error

All error responses include a detail field with a human-readable message that is displayed directly in the upload zone UI.

`POST /api/tailor`

(Endpoint is live, UI wiring in progress)

Performs gap analysis between a resume and a job description using Gemini.

Request body:

{
  "resume_data": { ... },
  "job_description": "We are looking for..."
}

Response:

{
  "analysis": {
    "missing_skills": ["Docker", "Kubernetes"],
    "suggestions": "...",
    "improved_summary": "...",
    "improved_bullets": [{ "exp_id": "exp1", "bullet_idx": 0, "new_text": "..." }]
  }
}

`POST /api/export/pdf`

Generates a continuous, non-paginated PDF using a headless Chromium instance via Playwright.

Request body:

{
  "html": "<!DOCTYPE html><html>...</html>"
}

Response: Binary PDF stream.

`POST /api/export/docx`

Generates a perfectly styled Word Document via python-docx.

Request body:

{
  "bio": { ... },
  "experience": [ ... ],
  "education": [ ... ],
  "skills": [ ... ]
}

Response: Binary DOCX stream.

Privacy & Data Model

CV Modernizer is designed to minimise data retention.

What is stored (Supabase `cv_cache`)

Column	Example	Purpose
`email`	`user@example.com`	Deduplication key (UNIQUE)
`phone`	`+91 9999999999`	Secondary identity signal
`name`	`Jane Doe`	Display / debugging only
`experiences`	`[{company, role, startDate, endDate}]`	Metadata — no bullet points
`education`	`[{institution, degree, startDate, endDate}]`	Metadata only
`created_at`	`2026-04-19T05:00:00Z`	Immutable first-upload timestamp
`updated_at`	`2026-04-19T05:00:00Z`	TTL reference — updated on each re-upload

What is never stored

The original PDF/DOCX file
Resume bullet points / achievements
Skills lists
Links / URLs
AI-generated output

Rate limit logic

Upload request
  → extract email from raw text
  → query Supabase: SELECT updated_at WHERE email = ? AND updated_at >= NOW() - 3h
  → if row found: reject 429 with "try again in Xh Ym"
  → if not found: proceed → Gemini → INSERT/UPDATE cache

The TTL window is based on updated_at (not created_at), so re-uploading always resets the 3-hour clock.

Upload Guards

The backend will reject an upload with 422 if:

No valid email address is found in the extracted text
No phone number is found in the extracted text

This prevents spam uploads and ensures the rate-limiting system has a reliable deduplication key.

Resume Templates

Five templates are rendered purely in CSS/HTML — no image assets required.

ID	Name	Style
`minimalist`	Minimalist	Clean single-column, generous whitespace
`modern-tech`	Modern Tech	Split layout — skills/contact sidebar + experience main
`executive`	Executive	Bold header bar, formal section dividers
`creative`	Creative	Accent left border, colourful section headers
`elegant`	Elegant	Serif-inspired, centred header, refined typography

Each template respects the selected primary colour from 12 available palettes (Slate, Navy, Forest, Burgundy, Gold, Plum, Teal, Copper, Rose, Midnight, Sage, Steel).

Known Limitations

Area	Limitation
Gemini quota	Free-tier keys have request-per-minute limits. Heavy usage will hit `503`. Retry after ~60 seconds.
Scanned PDFs	Image-based PDFs (scans with no selectable text) will return no extracted text. Use a text-based PDF.
Phone detection	The regex targets common international formats. Unusual formatting (e.g. spaces instead of dashes) may not be detected.
Export Height	PDFs are generated as a single continuous sheet. If you want paginated A4 outputs, custom print margins need to be injected into the Playwright engine.
Ko-fi payments	Ko-fi's checkout (Stripe) cannot run inside an iframe or restricted popup. The Ko-fi button opens a standard browser popup.
CORS	Backend CORS allows `localhost:3000` and `localhost:3001` only. Update `allow_origins` in `main.py` before deploying.

Support the Project

If this tool saved you time, consider buying a coffee:

☕ Ko-fi (international): ko-fi.com/yourhandle
🇮🇳 UPI (India): yourusername@upi

Both options are available directly in the sidebar of the app.

License

This project is licensed under the MIT License. See the LICENSE file for details.

Name		Name	Last commit message	Last commit date
backend		backend
docs		docs
frontend		frontend
.dockerignore		.dockerignore
.gitignore		.gitignore
CNAME		CNAME
DEPLOYMENT.md		DEPLOYMENT.md
LICENSE		LICENSE
README.md		README.md
deploy.sh		deploy.sh
docker-compose.yml		docker-compose.yml

Latest commit

History

Folders and files

Repository files navigation

CV Modernizer

Table of Contents

Overview

Screenshots

🏠 Landing Page

✏️ Live Side-by-Side Editor

Key Features

Architecture

Tech Stack

Frontend

Backend

Infrastructure

Project Structure

Getting Started

Prerequisites

1. One-Click VPS Deploy (Recommended)

2. Manual Setup (Local Development)

Prerequisites

Backend Setup

Frontend Setup

Environment Variables

API Reference

GET /api/health

POST /api/extract

POST /api/tailor

POST /api/export/pdf

POST /api/export/docx

Privacy & Data Model

What is stored (Supabase cv_cache)

What is never stored

Rate limit logic

Upload Guards

Resume Templates

Known Limitations

Support the Project

License

About

Topics

Resources

Stars

Watchers

Forks

Releases

Packages

Contributors

Languages

`GET /api/health`

`POST /api/extract`

`POST /api/tailor`

`POST /api/export/pdf`

`POST /api/export/docx`

What is stored (Supabase `cv_cache`)