2. 🎯 Semantic Job Description Matcher

• TF-IDF Vectorization: Transforms resume and target job descriptions into a vector space of word importance scores.
• Hybrid Match Scorer: Combines cosine similarity (50%) and physical skill token coverage (50%) to prevent false-positive matching on filler words while maintaining deep semantic relevance.
• Gap Analysis & Badges: Categorizes job keywords into Matching Skills Possessed (green) and Missing Skills (Skill Gap) (red) for instant visualization.

3. 💼 Intelligent Career Planner & Skill Gap Roadmaps

• Category Prediction: Predicts the primary resume profile (e.g., Backend Development, Data Science & AI, DevOps & Cloud, etc.) based on parsed skill concentrations.
• Market Insights: Displays estimated salaries, growth trends (e.g., +35% Exponential Growth), and high-demand skills for matched career tracks.
• Interactive Roadmaps: For every missing career skill, the engine suggests curated next-steps, recommended courses, and topics to study.

4. 📂 Reference Skill Taxonomy Browser

• Displays all 200+ technology keywords tracked by the custom Regex engine, divided into 8 distinct technical folders (Programming Languages, Web Development, Databases, Cloud & DevOps, AI/ML, Mobile & Systems, Software Engineering Concepts, and Soft Skills).

5. 📥 Professional PDF Report Export

• Generates and downloads a beautifully styled, print-ready PDF analysis report using ReportLab containing the full compatibility scorecard, actionable items, and career roadmap highlights.

📂 Project Architecture

AI_Resume_Project/
│
├── app.py                      # Flask main server (routing, middleware, APIs)
├── requirements.txt            # Python dependencies
├── README.md                   # Installation & documentation (This file)
│
├── templates/
│   └── index.html              # Custom HTML5 dashboard landing layout (Inter & Outfit fonts)
│
├── static/
│   ├── css/
│   │   └── styles.css          # Sleek glassmorphism stylesheets
│   └── js/
│       └── main.js             # Client SPA controller, fetch requests, & gauge logic
│
├── data/
│   └── sample_jds.json         # Predefined high-quality Job Descriptions for testing
│
└── utils/
    ├── parser.py               # Extract text and page count from PDF resumes
    ├── preprocessing.py        # Tokenization, cleaning, and regex contact/name parsing
    ├── skill_extractor.py      # Skill taxonomies & custom Regex-based matchers
    ├── ats_scorer.py           # ATS scoring calculations, action verbs, and thresholds
    ├── matcher.py              # TF-IDF + Cosine Similarity matching engine
    ├── recommender.py          # Career mapping, skill gap analysis, and course paths
    ├── pdf_generator.py        # ReportLab PDF print-out layout manager
    └── visualizer.py           # [Legacy] Streamlit dashboard visualization utilities

🛠️ Installation & Setup

Follow these steps to run the application locally on your computer:

Prerequisites

Make sure you have Python 3.8+ installed.

1. Navigate to the Project Directory

Open your terminal and navigate to the project workspace:

cd c:/Users/paras/OneDrive/Documents/AI_Resume_Project

2. Install Dependencies

Install all required libraries using pip:

pip install -r requirements.txt

(Note: The application will automatically download the spaCy en_core_web_sm model on first launch. If you face any issues, you can run python -m spacy download en_core_web_sm manually).

3. Launch the Flask Server

Start the local server:

python app.py

4. Open in Browser

Once launched, open your web browser and navigate to:

http://127.0.0.1:5000

🚀 How to Test the System

1. Step 1: Get Scored

   • Navigate to the ATS Scorecard tab.
   • Toggle the Student / Fresher option if you are a student or fresh graduate.
   • Upload any sample tech resume (in .pdf format) inside the upload box.
   • Click Analyze Uploaded Resume.
   • Review your overall compatibility gauge, extracted contact information, and skill distribution folders.
   • Click Export Detailed PDF Report to download a print-ready PDF scorecard.

2. Step 2: Match against a Job Description (JD)

   • Navigate to the Job Matcher Portal tab.
   • Select one of the pre-loaded jobs from the dropdown (e.g. DeepMinded AI Labs - AI/ML Engineer or TechNova Solutions - Backend Software Engineer).
   • Alternatively, paste any job description from LinkedIn, Indeed, or elsewhere.
   • Click Run Compatibility Comparison to view matching/missing badges and semantic similarity metrics.

3. Step 3: Roadmap Upskilling

   • Go to the Career Roadmap tab.
   • Check your primary predicted category.
   • Expand the recommended careers to view tailored roadmap milestones and links to master missing skills.

🧑‍💻 Technical Highlights (For Portfolio/Resume)

• Hybrid Matcher Architecture: Developed a specialized scoring metric combining structural text representation (TF-IDF bag-of-words vectorization) with discrete set theory logic (Hard Skill token intersection ratio).
• Word-Boundary Symbol Parser: Crafted case-insensitive Regular Expressions to accurately detect specialized programming syntax (C++, C#, .NET, Node.js) while maintaining strict word boundary safety to eliminate substring collisions and bullet indexing errors.
• Client-Side SPA Architecture: Implemented a central client-side state object (const state = {...}) in vanilla JavaScript, ensuring parsed resume metadata, matched job results, and navigation tab views persist seamlessly without full-page reloads.
• Dynamic PDF Streaming: Configured an on-the-fly binary buffer using io.BytesIO and ReportLab's SimpleDocTemplate to compile and stream custom-branded PDF scorecards directly to client browsers without saving files to server disks.