How bajajharsha/recruitment-process-automation-tool works
Project Type & Purpose
• This is a full-stack application aimed at automating parts of the recruitment process. Its functionality is twofold: matching resumes to job descriptions and analyzing interview video content for insights.
• The backend (API services) focuses on processing and analysis (e.g., text extraction, NLP, video analysis, similarity search using vector stores), while the frontend is a Streamlit-based user interface to facilitate user interactions with these features.
Analyzing the File Structure
• Top-Level Directories:
• “StreamlitApp”: Contains scripts like AnalyzeVideo.py, JobMatchingSystem.py, home.py, and streamlit_app.py that define the web interface where users interact with the tool.
• Configuration & Environment:
• The .env file holds environment variables including API keys for external services.
• The README and requirements.txt indicate dependencies like fastapi/uvicorn (for backend API services) and streamlit (for the UI), as well as several third-party services (e.g., AssemblyAI, Pinecone, Google Generative AI).
Key Components and Their Interactions
• Frontend Component:
• Represented by the StreamlitApp folder.
• Responsible for user interactions (uploading resumes, job descriptions, videos).
• Initiates calls to the backend API for processing analysis.
• Backend/API Component:
• Contained in the Api directory.
• Comprises multiple functional modules:
▪ Video analysis, text extraction, NLP analysis, similarity search and vector store management.
▪ The file app.py likely wraps up these functionalities and runs a FastAPI server with uvicorn to serve API requests.
• External Services & Dependencies:
• External APIs and libraries are integrated for language processing (e.g., langchain, sentence-transformers), video processing (moviepy), and cloud-based analysis (AssemblyAI, Google Generative AI, Pinecone).
Detailed Elements
• Each component should be clearly labeled (e.g., “Streamlit Frontend”, “Resume Matching API”, “Video Analysis API”, “NLP Processing Module”).
• User interactions: uploads resume, job description, video → StreamlitApp.
• StreamlitApp making HTTP calls to the API endpoints in the Api folder.
• API modules processing data and then reaching out to external APIs for advanced analysis.
• Adding brief notes on each component to explain its role.