# AI Resume Intelligence Analyzer A **production-grade, fully dynamic** Python terminal tool that evaluates a candidate's resume against a target job role and optional company. ## Features - **Real Internet Scraping** — fetches live job requirements via DuckDuckGo (no API key required) - **24-Hour Cache** — avoids re-scraping on every run - **Skill Normalization** — `sklearn` → `scikit-learn`, `k8s` → `kubernetes`, etc. - **5-Feature ML Scoring** — scikit-learn Linear Regression model with MinMaxScaler - **Logistic Probability Estimation** — shortlisting probability with confidence band - **Professional Competency Analysis** — 8-axis personality spider chart - **Dark-Themed Visual Charts** — Skill Radar + Competency Spider (matplotlib) - **Week-by-Week Learning Roadmap** — 30+ skill knowledge base - **Free Resource Links** — YouTube, freeCodeCamp, official docs, GitHub awesome repos - **Resume Bullet Suggestions** — action-verb improvements for missing skills - **Categorized Interview Questions** — Conceptual / Coding / Scenario / Behavioral - **Full Logging** — `logs/analyzer.log` for debugging --- ## Project Structure ``` resume_ai_analyzer/ ├── main.py # Entry point — 12-step pipeline ├── requirements.txt ├── README.md ├── modules/ │ ├── logger.py # Centralized logging │ ├── utils.py # Shared helpers │ ├── cache_manager.py # 24h on-disk cache │ ├── resume_parser.py # PyPDF2 + pdfminer.six fallback │ ├── skill_extractor.py # Regex + DB + keyword density + exp signal │ ├── skill_normalizer.py # Alias normalization │ ├── internet_job_scraper.py # DuckDuckGo multi-source scraper │ ├── skill_gap_analyzer.py # Gap analysis + priority scoring │ ├── skill_vectorizer.py # One-hot vectors + cosine similarity │ ├── regression_model.py # 5-feature Linear Regression │ ├── probability_estimator.py # Logistic probability + confidence band │ ├── personality_analyzer.py # 8-axis keyword-signal competency │ ├── radar_chart_generator.py # Skill coverage radar chart │ ├── spider_chart_generator.py # Professional competency spider chart │ ├── learning_roadmap_generator.py # Week-by-week roadmap │ ├── resource_finder.py # Free learning resource links │ ├── resume_improvement_advisor.py # Resume bullet suggestions │ ├── interview_question_generator.py # Interview questions │ └── report_generator.py # Full terminal + file report ├── data/ │ ├── skill_database.json # 200+ skills with categories │ ├── skill_aliases.json # 100+ alias normalizations │ └── fallback_role_skills.json # 17 roles for offline fallback ├── cache/ │ └── role_skill_cache.json # Auto-populated cache ├── outputs/ │ ├── skill_radar_chart.png # Generated radar chart │ ├── personality_spider_chart.png # Generated spider chart │ └── analysis_report.txt # Full saved report └── logs/ └── analyzer.log # Detailed debug log ``` --- ## Installation ### 1. Python Backend ```bash pip install -r requirements.txt ``` ### 2. React Frontend ```bash cd frontend npm install cd .. ``` --- ## Usage ### Launching the Full SaaS Dashboard (Recommended) To run both the Python FastAPI backend and the React Vite frontend simultaneously, simply run: ```bash start.bat ``` *(This will automatically start the backend server, launch the frontend server, and open the beautiful React UI in your default web browser.)* ### Launching the Terminal CLI (Fallback) If you prefer the original command-line interface: ```bash python main.py ``` At the prompts: 1. **Paste resume text** or enter a **PDF/TXT file path** 2. Enter the **target job role** (e.g. `Machine Learning Engineer`) 3. Enter the **company name** (optional, e.g. `Google`) 4. Type `exit` at any prompt to quit. --- ## Output ``` outputs/ ├── skill_radar_chart.png ├── personality_spider_chart.png └── analysis_report.txt ``` --- ## Notes - First run requires internet access for scraping; results are cached for 24 hours - PDF extraction uses PyPDF2 with pdfminer.six as fallback - All outputs are dynamic — results change with different resumes, roles, and companies # AIresumeanalyzer # AIRC