██████╗ ███████╗███████╗██╗   ██╗███╗   ███╗███████╗
██╔══██╗██╔════╝██╔════╝██║   ██║████╗ ████║██╔════╝
██████╔╝█████╗  ███████╗██║   ██║██╔████╔██║█████╗  
██╔══██╗██╔══╝  ╚════██║██║   ██║██║╚██╔╝██║██╔══╝  
██║  ██║███████╗███████║╚██████╔╝██║ ╚═╝ ██║███████╗
╚═╝  ╚═╝╚══════╝╚══════╝ ╚═════╝ ╚═╝     ╚═╝╚══════╝
██████╗  ██████╗  █████╗ ███████╗████████╗███████╗██████╗ 
██╔══██╗██╔═══██╗██╔══██╗██╔════╝╚══██╔══╝██╔════╝██╔══██╗
██████╔╝██║   ██║███████║███████╗   ██║   █████╗  ██████╔╝
██╔══██╗██║   ██║██╔══██║╚════██║   ██║   ██╔══╝  ██╔══██╗
██║  ██║╚██████╔╝██║  ██║███████║   ██║   ███████╗██║  ██║
╚═╝  ╚═╝ ╚═════╝ ╚═╝  ╚═╝╚══════╝   ╚═╝   ╚══════╝╚═╝  ╚═╝

Resume Roaster is a fully functional ATS (Applicant Tracking System) simulator that tears apart your resume, matches it against real job descriptions, finds the gaps, and tells you exactly why you're getting ghosted — before a recruiter does.

🚀 Demo · 📦 Installation · 🧠 How It Works · ⚙️ Architecture · 📊 Features · 🤝 Contributing

---

🎯 The Problem

Every year, millions of qualified candidates get rejected before a human ever sees their resume — filtered out silently by ATS systems they don't understand.

Students and early-career professionals face a brutal reality:

• 📭 Applied to 100+ jobs, heard back from 3?
• 🤷 Don't know why you're getting rejected?
• 🧩 Not sure what keywords or skills are missing?
• 🕵️ Never seen the inside of an ATS before?

Resume Roaster changes that. It puts you on the other side of the system.

---

💡 What It Does

Resume Roaster simulates a real ATS pipeline end-to-end:

Step	What Happens
📄 Parse	Extracts structured data from your PDF resume
🧹 Clean	Normalizes and preprocesses raw text
🔢 Score	Runs a weighted ML-inspired scoring engine
🧠 Match	Compares your resume to a job description using TF-IDF + Cosine Similarity
🔍 Gap Analysis	Identifies matched vs. missing keywords
🎯 Decide	Simulates a recruiter's shortlisting decision
💬 Feedback	Gives actionable, human-readable suggestions
📥 Report	Generates a downloadable analysis report

---

⚙️ System Architecture

                        ┌─────────────────────┐
                        │     PDF Resume       │
                        └─────────┬───────────┘
                                  │
                        ┌─────────▼───────────┐
                        │   Parser Module      │  ← PyPDF2 + Regex + Section Detection
                        └─────────┬───────────┘
                                  │
                        ┌─────────▼───────────┐
                        │  Feature Engineering │  ← Projects, Skills, Experience, etc.
                        └─────────┬───────────┘
                                  │
               ┌──────────────────┼──────────────────┐
               │                  │                  │
    ┌──────────▼────────┐ ┌───────▼────────┐ ┌──────▼──────────┐
    │  ML Scoring Engine│ │ NLP JD Matcher  │ │ Keyword Gap     │
    │  (Weighted Score) │ │ (TF-IDF +       │ │ Analyzer        │
    │                   │ │  Cosine Sim)    │ │ (Set Ops)       │
    └──────────┬────────┘ └───────┬────────┘ └──────┬──────────┘
               │                  │                  │
               └──────────────────▼──────────────────┘
                                  │
                        ┌─────────▼───────────┐
                        │  Decision Engine     │  ← Shortlist: YES / MAYBE / NO
                        └─────────┬───────────┘
                                  │
                        ┌─────────▼───────────┐
                        │  Feedback + Report   │  ← Strengths, Gaps, Suggestions
                        └─────────┬───────────┘
                                  │
                        ┌─────────▼───────────┐
                        │   Streamlit UI       │  ← Interactive Web Interface
                        └─────────────────────┘

---

📊 Features

🧩 Core Modules

📁 parser.py — Resume Parser

• Extracts raw text from PDF using PyPDF2
• Cleans and normalizes text (lowercasing, punctuation, whitespace)
• Detects resume sections (Projects, Skills, Experience, Education, etc.)
• Extracts structured features via regex and keyword detection

Extracted Feature Set:

{
  "projects":        [...],
  "skills":          [...],
  "links":           [...],
  "achievements":    [...],
  "experience":      [...],
  "certifications":  [...]
}

📊 scorer.py — ML Scoring Engine

Simulates ATS scoring with a weighted model that penalizes fake inflation.

Feature	Weight
Projects	30%
Experience	25%
Skills	20%
Achievements	15%
Certifications/Links	10%

• Strict penalty system — no inflated scores
• Score normalization
• Capped final score to prevent false positives

🧠 nlp_matcher.py — JD Matching

Uses classical NLP to compare your resume against a job description:

TF-IDF Vectorization → Cosine Similarity → Match %

• Handles varied JD formats
• Language-agnostic keyword weighting
• Outputs similarity percentage (0–100%)

🔍 gap_analyzer.py — Keyword Gap Analyzer

matched  = resume_keywords ∩ jd_keywords   # ✅ You have these
missing  = jd_keywords - resume_keywords   # ❌ You're missing these

Pinpoints exactly which skills/tools to add to pass the ATS filter.

🎯 decision_engine.py — Recruiter Simulation

Resume Score (60%) + JD Match (40%) → Shortlist Decision

Result:
  ✅ Shortlisted: YES     → Score ≥ 75 & Match ≥ 70
  ⚠️ Shortlisted: MAYBE  → Borderline signals
  ❌ Shortlisted: NO      → Below threshold

🧠 skill_recommender.py — Skill Gap Recommender

Maps missing keywords to actionable learning suggestions:

Missing: docker   →  "Learn Docker for containerized deployments"
Missing: aws      →  "Get AWS Cloud Practitioner certified"
Missing: sql      →  "Practice SQL on LeetCode / HackerRank"

📊 visualizer.py — Visual Analytics

Generates Matplotlib charts showing:

• Resume strength breakdown (bar chart by feature)
• ATS score vs. JD match comparison
• Keyword coverage heatmap

---

🖥️ UI Preview

┌─────────────────────────────────────────────────────┐
│  🔥 RESUME ROASTER                          v1.0     │
├─────────────────────────────────────────────────────┤
│  📄 Upload Resume (PDF)     🎯 Select Target Role    │
│  ┌─────────────────────┐    ┌─────────────────────┐  │
│  │  resume.pdf  ✅      │    │  Backend Engineer ▾ │  │
│  └─────────────────────┘    └─────────────────────┘  │
│                                                       │
│  📋 Paste Job Description                            │
│  ┌─────────────────────────────────────────────────┐ │
│  │  We are looking for a Python developer with...  │ │
│  └─────────────────────────────────────────────────┘ │
│                                                       │
│              [ 🔥 ROAST MY RESUME ]                  │
├─────────────────────────────────────────────────────┤
│  RESUME SCORE: ████████░░  78%                       │
│  JD MATCH:     ██████░░░░  65%                       │
│  DECISION:     ⚠️  MAYBE  (Confidence: 70%)           │
│                                                       │
│  ✅ python  ✅ backend   ❌ docker  ❌ aws            │
└─────────────────────────────────────────────────────┘

---

📦 Installation

Prerequisites

• Python 3.10+
• pip

Setup

# 1. Clone the repository
git clone https://github.com/yourusername/resume-roaster.git
cd resume-roaster

# 2. Create a virtual environment
python -m venv venv
source venv/bin/activate        # On Windows: venv\Scripts\activate

# 3. Install dependencies
pip install -r requirements.txt

# 4. Run the app
streamlit run app.py

Dependencies

streamlit
PyPDF2
scikit-learn
matplotlib
pandas
numpy
re

---

🚀 Demo

Sample Output

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
        🔥 RESUME ROAST RESULTS
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

📊 RESUME SCORE     :  78 / 100
🎯 JD MATCH         :  65%
🤖 ATS DECISION     :  ⚠️  MAYBE
📈 CONFIDENCE       :  70%

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
✅ MATCHED KEYWORDS
   ✔ python   ✔ backend   ✔ rest api

❌ MISSING KEYWORDS
   ✘ docker   ✘ aws   ✘ kubernetes

💡 SKILL SUGGESTIONS
   → Learn Docker for containerized apps
   → Get AWS Cloud Practitioner certified
   → Explore Kubernetes basics on KodeKloud

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
💬 FEEDBACK
   ✔ Strong project section
   ✔ Good use of technical keywords
   ⚠ Add internship / work experience
   ⚠ Include deployment-related skills
   💥 Power-up: Add a GitHub projects link
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

---

🧠 Concepts Demonstrated

Domain	Concepts
🤖 Machine Learning	Feature engineering, weighted scoring, model-based thinking
📚 NLP	TF-IDF vectorization, cosine similarity, keyword extraction
🔍 Information Retrieval	Keyword matching, relevance scoring, document similarity
⚙️ Software Engineering	Modular architecture, separation of concerns, reusable components
📊 Data Science	Feature normalization, score distribution, visualization
🧠 System Design	End-to-end pipeline design, ATS simulation, decision logic

---

🗂️ Project Structure

resume-roaster/
│
├── app.py                  # Streamlit entry point
│
├── modules/
│   ├── parser.py           # PDF text extraction + feature engineering
│   ├── scorer.py           # Weighted ML scoring engine
│   ├── nlp_matcher.py      # TF-IDF + Cosine Similarity JD matcher
│   ├── gap_analyzer.py     # Keyword gap detection
│   ├── decision_engine.py  # Recruiter shortlist simulation
│   ├── skill_recommender.py# Skill gap → learning suggestions
│   ├── visualizer.py       # Matplotlib charts
│   ├── feedback_engine.py  # Human-readable feedback generator
│   └── report_generator.py # Downloadable report export
│
├── data/
│   └── skill_map.json      # Skill → suggestion mapping
│
├── assets/
│   └── sample_resume.pdf   # Test resume
│
├── requirements.txt
├── LICENSE
└── README.md

---

⚠️ Limitations & Future Work

Current Limitations

• Rule-based feature extraction (no deep semantic parsing)
• Synthetic scoring model (not trained on real recruiter data)
• Basic keyword matching (no synonym/contextual awareness)
• No deep ML model training

🔭 Roadmap

• 🤗 BERT/Sentence-Transformers for semantic JD matching
• 🧠 Train on real recruiter feedback data
• 🌍 Multi-language resume support
• 📊 Dashboard analytics across multiple resumes
• 🔗 LinkedIn profile import
• 🧾 LaTeX resume generation from feedback

---

🤝 Contributing

Contributions are welcome! Here's how:

# Fork the repo, then:
git checkout -b feature/your-feature-name
git commit -m "feat: add your feature"
git push origin feature/your-feature-name
# Open a Pull Request 🎉

Please follow the Contributor Guidelines and ensure all new modules include docstrings and unit tests.

---

📄 License

This project is licensed under the MIT License — see LICENSE for details.

---

🙌 Acknowledgements

• Streamlit — for the rapid UI layer
• scikit-learn — TF-IDF and vectorization tools
• PyPDF2 — PDF parsing
• Matplotlib — visualization

---

Built with 🔥 to help students stop getting ghosted.

If this helped you land an interview — drop a ⭐ on the repo.