Transform your hiring process with cutting-edge AI/ML technology

HireScope is an advanced resume analysis platform that leverages state-of-the-art NLP, semantic embeddings, and machine learning to revolutionize how you evaluate candidates. Built with modern web technologies and AI models, it provides instant, actionable insights to help both job seekers optimize their resumes and recruiters find the perfect match.

• Features
• Demo
• Technology Stack
• Architecture
• Installation
• Quick Start
• Contact

• Multi-Dimensional Scoring: Evaluates resumes across 5 key dimensions
  • Structural Quality (sections, formatting)
  • Keyword Match (job description alignment)
  • Semantic Relevance (contextual understanding)
  • Readability (clarity and conciseness)
  • Tone & Impact (action verbs, achievements)
• Role Detection: Automatically identifies candidate profile (Tech, Manager, Creative, General)
• Semantic Embeddings: Uses sentence-transformers for deep contextual understanding (384-dimensional vectors)
• Weighted Keyword Matching: Prioritizes required (2x), preferred (1x), and bonus (0.5x) keywords

• ATS Compatibility Scoring: 0-100 score with specific formatting issues
• Gap Analysis: Identifies missing skills with priority levels (Critical → High → Medium)
• Competitive Intelligence: Percentile ranking and skill strength assessment
• Quick Wins Identification: 3-5 high-impact, easy-to-implement improvements
• Before/After Examples: Concrete improvement suggestions
• Time-to-Implement Estimates: Helps prioritize resume updates

• 7 Specialized Analysis Modules:
  1. ATS Optimization
  2. Gap Analysis
  3. Competitive Benchmarking
  4. Section-Specific Recommendations
  5. Keyword Optimization
  6. Format & Structure
  7. Content Enhancement
• 60+ Expert Tips: Enhanced RAG knowledge base (6x industry best practices)
• Multi-Level Prioritization: Critical → High → Medium → Optional
• Overall Grade: A/B/C/D rating with detailed breakdown

• Fast Processing: 2-3 second average analysis time
• File Format Support: PDF and DOCX
• OCR Capability: Handles scanned documents (optional Tesseract integration)
• Caching Layer: Redis + memory fallback for performance
• Lazy Loading: 92% faster startup with on-demand model loading
• Error Handling: Comprehensive error messages and recovery
• CORS Support: Cross-origin resource sharing enabled

┌─────────────────┐      ┌──────────────────┐      ┌────────────────────┐
│  Upload Resume  │ ───> │  AI Processing   │ ───> │  Results Dashboard │
│  + Job Desc     │      │  (2-3 seconds)   │      │  + Insights        │
└─────────────────┘      └──────────────────┘      └────────────────────┘

• Framework: FastAPI 0.115.4 - High-performance Python web framework
• Server: Uvicorn - Lightning-fast ASGI server
• AI/ML Stack:
  • Sentence Transformers 2.2.2 - Semantic embeddings
  • spaCy 3.8.7 - NLP preprocessing
  • Transformers 4.57.1 - Model inference
  • scikit-learn - Scoring algorithms
• Document Processing:
  • PyPDF2 - PDF extraction
  • python-docx - DOCX parsing
  • pytesseract - OCR (optional)
• Caching: Redis + in-memory fallback
• Validation: Pydantic 2.0+

• Framework: React 19.1.1 - Latest with concurrent features
• Build Tool: Vite 7.1.7 - Next-generation frontend tooling
• Styling: Tailwind CSS 3.4.18 - Utility-first CSS
• Animations: Framer Motion 12.23.24 - Production-ready animations
• Charts: Recharts 3.3.0 - React charting library
• Icons: Heroicons 2.2.0 - Beautiful SVG icons
• HTTP Client: Axios 1.13.1 - Promise-based requests
• UI Components: Headless UI 2.2.9 - Unstyled components

• Embeddings: all-MiniLM-L6-v2 (384-dimensional)
• NLP: spaCy en_core_web_sm English model
• Tokenization: Sentence-level semantic chunking

┌──────────────────────────────────────────────────────────────┐
│                         Frontend (React)                      │
│  ┌─────────────┐  ┌──────────────┐  ┌──────────────────┐   │
│  │   Hero      │  │  UploadZone  │  │ ResultsDashboard │   │
│  │  Component  │  │  (Drag&Drop) │  │   (Analytics)    │   │
│  └─────────────┘  └──────────────┘  └──────────────────┘   │
└──────────────────────────────────────────────────────────────┘
                              │
                              │ HTTP/REST API
                              │
┌──────────────────────────────────────────────────────────────┐
│                      Backend (FastAPI)                        │
│  ┌─────────────────────────────────────────────────────┐    │
│  │                  API Routes Layer                    │    │
│  │  POST /api/resume/upload                             │    │
│  │  POST /api/resume/analyze                            │    │
│  │  GET  /api/health                                    │    │
│  └───────���─────────────────────────────────────────────┘    │
│                              │                                │
│  ┌─────────────────────────────────────────────────────┐    │
│  │              Core Processing Pipeline                │    │
│  │  1. Parsing      → Extract text (PDF/DOCX/OCR)      │    │
│  │  2. Preprocessing → spaCy tokenization/segmentation  │    │
│  │  3. Embeddings   → SentenceTransformer vectors       │    │
│  │  4. Keywords     → NLP extraction + matching         │    │
│  │  5. Scoring      → Multi-dimensional algorithm       │    │
│  │  6. Insights     → RAG-enhanced recommendations      │    │
│  │  7. Response     → Structured JSON + metadata        │    │
│  └─────────────────────────────────────────────────────┘    │
│                              │                                │
│  ┌─────────────────────────────────────────────────────┐    │
│  │                  AI/ML Layer                         │    │
│  │  ┌──────────────┐  ┌────────────┐  ┌────────────┐  │    │
│  │  │ Embeddings   │  │   spaCy    │  │ scikit-    │  │    │
│  │  │ Model Cache  │  │  NLP Model │  │  learn     │  │    │
│  │  └──────────────┘  └────────────┘  └────────────┘  │    │
│  └─────────────────────────────────────────────────────┘    │
└──────────────────────────────────────────────────────────────┘
                              │
                              │
                    ┌─────────────────┐
                    │  Redis Cache    │
                    │  (Optional)     │
                    └─────────────────┘

1. Upload Phase

   User → File Selection → Validation → FormData → API Request
   

2. Processing Phase

   API → Parse Document → Extract Text → Clean & Tokenize
       → Generate Embeddings → Match Keywords → Calculate Scores
       → Generate Insights → Build Response
   

3. Presentation Phase

   API Response → Frontend State → Animate Components
               → Render Charts → Display Insights
   

• Python: 3.13+ (Download)
• Node.js: 18+ (Download)
• npm: 10+ (comes with Node.js)
• Git: Latest version (Download)

• Tesseract: OCR engine (Installation Guide)
• Poppler: PDF rendering library (Installation Guide)

git clone https://github.com/ParthSharma272/HireScope.git
cd HireScope

cd backend

python3 -m venv venv
source venv/bin/activate
pip install --upgrade pip

pip install -r requirements.txt
python -m spacy download en_core_web_sm

cd ../hirescope-frontend
npm install

cd backend
source venv/bin/activate
uvicorn app:app --reload --host 0.0.0.0 --port 8000

Output:

INFO:     Uvicorn running on http://0.0.0.0:8000
INFO:     Application startup complete.

# In a new terminal window
cd hirescope-frontend
npm run dev

Output:

VITE v7.1.7  ready in 328 ms

➜  Local:   http://localhost:5175/
➜  Network: use --host to expose

Open your browser and navigate to:

• Frontend: http://localhost:5175
• Backend API: http://127.0.0.1:8000
• API Docs: http://127.0.0.1:8000/docs (Interactive Swagger UI)
• Health Check: http://127.0.0.1:8000/api/health

1. Upload Resume: Drag and drop a PDF or DOCX file
2. Paste Job Description: Copy-paste the job posting text
3. Click "Analyze Resume": Wait 2-3 seconds for AI processing
4. View Results: Explore scores, keywords, and insights
5. Implement Suggestions: Apply AI-powered recommendations

Parth Sharma

• 📧 Email: parthsharma23212@gmail.com
• 💼 LinkedIn: linkedin.com/in/parth-sharma-08b1b424b
• 🐙 GitHub: @ParthSharma272
• 📝 Medium: @parthsharma23212
• 🌐 Portfolio: portfolio-website-f311.vercel.app

If you find HireScope helpful, please consider:

• ⭐ Star this repository
• 🐛 Report bugs and suggest features via Issues
• 🤝 Contribute code or documentation
• 💬 Share with your network