AI Resume Analyzer

An AI-powered web app that analyzes a resume and gives clear, useful feedback: strengths, weaknesses, missing skills, and suggestions for improvement.

Built as a beginner-friendly MVP using LangChain, FastAPI, Groq and Next.js.

Features

• Upload your resume (PDF or Word .docx) and get an instant AI analysis
• ATS score, section-wise scores, strengths, weaknesses and suggestions
• Job match percentage and missing keywords (when you paste a job description)
• Feedback in English, Roman Urdu, Urdu or Hindi
• Auto-improve: AI finds the weakest lines in your resume and shows a before/after rewrite for each
• Download improved CV: AI rewrites your whole resume and gives you an editable Word (.docx) file or a PDF
• Bullet rewriter: paste a weak resume line and get 3 stronger versions
• Cover letter generator: a tailored cover letter from your resume + job description
• Skill gap roadmap: the skills you are missing for a target role, with time estimates, learning steps and free resources
• History: past analyses and roadmaps are saved in your browser, with a chart of how your ATS score improves
• Clean, modern web interface
• Fast responses powered by Groq's free cloud LLM API

Tech Stack

Layer	Technology
Frontend	Next.js, React
Backend	Python, FastAPI
AI	LangChain + Groq API

Project Structure

resume-analyzer/
├── backend/     # FastAPI server + LangChain logic
├── frontend/    # Next.js user interface
└── README.md

Getting Started

Prerequisites

• Python 3.10+
• Node.js 18+
• A free API key from Groq

1. Clone the repository

git clone https://github.com/SyedMuhammdSohaib/AI-ResumeAnalyzer.git
cd AI-ResumeAnalyzer

2. Set up the backend

cd backend
python -m venv venv
# Windows
venv\Scripts\activate
# Mac/Linux
source venv/bin/activate

pip install -r requirements.txt
cp .env.example .env

Edit backend/.env to fill in your keys.

If the model gives a 404 for your key, set GROQ_MODEL to openai/gpt-oss-120b or llama-3.3-70b-versatile.

Start the server:

uvicorn app.main:app --reload

The API runs at http://localhost:8000.

3. Set up the frontend

Open a new terminal:

cd frontend
npm install
cp .env.example .env.local  # or set NEXT_PUBLIC_API_URL in your shell
npm run dev

The app runs at http://localhost:3000.

Windows shortcut: double-click start-resume-analyzer.bat in the project root. It starts both servers and opens the browser.

4. Quick API check (optional)

With the backend running:

cd backend
pip install httpx
python test_analyze.py

How It Works

1. The user uploads a resume from the Next.js frontend.
2. The FastAPI backend extracts the text from the file.
3. LangChain sends the text to a Groq-hosted LLM with a carefully written prompt.
4. The analysis is returned to the frontend and displayed to the user.

API endpoints

Method	Path	What it does
GET	/health	Health check
POST	/analyze	Form: file (PDF or DOCX), optional job_description, language
POST	/rewrite	JSON: { "text": "...", "job_description": "..." } → 3 rewrites
POST	/improve	Form: file (PDF or DOCX), optional job_description → weak lines + fixes
POST	/improved-cv	Form: file (PDF or DOCX), optional job_description, rewrites, format (docx or pdf) → file
POST	/cover-letter	Form: file (PDF or DOCX), job_description → cover letter text
POST	/skill-roadmap	Form: file (PDF or DOCX), optional job_description, target_role, language → learning roadmap

Deployment

Backend (Vercel, free, no card needed)

The backend includes backend/api/index.py and backend/vercel.json, so it can run on Vercel as a Python serverless function.

1. In Vercel, click Add New → Project and import this repository.
2. Set Root Directory to backend and Framework Preset to Other.
3. Add environment variables: GROQ_API_KEY, GROQ_MODEL and ALLOWED_ORIGINS (see the table below).
4. Deploy, then open https://<your-backend>.vercel.app/health. It should return {"status":"ok"}.

Backend (Render, alternative)

1. Create a new Render Web Service and connect this repository.

2. Use the following settings:

   • Runtime: Python 3.12 (or compatible)
   • Build Command: pip install -r backend/requirements.txt
   • Start Command: uvicorn app.main:app --host 0.0.0.0 --port $PORT (Render sets $PORT; the included backend/Dockerfile also works if you prefer a Docker deploy)

3. Add the following environment variables under Environment:

   Variable	Description	Example
   GROQ_API_KEY	Your Groq API key	gsk_...
   GROQ_MODEL	LLM model to use	openai/gpt-oss-120b (or llama-3.3-70b-versatile)
   ALLOWED_ORIGINS	Comma-separated list of allowed frontend origins (CORS)	https://your-app.vercel.app,http://localhost:3000

4. Deploy the service. Your backend URL will be something like https://your-backend.onrender.com.

Frontend (Vercel)

1. Import this repository into Vercel.

2. Set the project root to frontend in Vercel project settings.

3. Add the following environment variable:

   Variable	Description	Example
   NEXT_PUBLIC_API_URL	Base URL of your deployed backend (no trailing slash)	https://your-backend.onrender.com

4. Deploy. The frontend will call ${NEXT_PUBLIC_API_URL}/analyze.

5. Update ALLOWED_ORIGINS on your Render backend to include your new Vercel URL.

Notes:

• Keep secrets out of version control. The repo already ignores .env and related files. See backend/.env.example and frontend/.env.example.
• CORS is read from ALLOWED_ORIGINS (comma-separated). By default it allows http://localhost:3000 for local development.
• The frontend reads the backend URL from process.env.NEXT_PUBLIC_API_URL (falls back to http://127.0.0.1:8000 for local development).

Future Improvements

• Interview question generator
• Download the analysis as a PDF report
• User accounts, so history syncs across devices

Author

Syed Muhammad Sohaib GitHub: @SyedMuhammdSohaib

License

This project is open source and available for learning purposes.