How taseskics/cvlens works
Project Purpose and Type
• CVLens is a full-stack web application that enables users to upload a resume and have its information parsed using a machine learning model.
• Its core components include a Next.js frontend for the user interface, a FastAPI backend for file handling and S3 integration, and AWS services (ECS Fargate, S3, API Gateway, and Lambda) that trigger machine learning model processing.
Main Components of the System
• Frontend (Next.js):
• Represents the user interaction layer.
• Includes pages for uploading resumes, viewing API documentation, and other static pages (e.g., "about-us").
• Backend (FastAPI in Python):
• Receives file uploads from the frontend.
• Processes the file by uploading it to AWS S3.
• Exposes endpoints (like /process-resume and /parsedData) for further data retrieval by the frontend.
• AWS S3 Bucket:
• Acts as the storage layer for uploaded resumes.
• Triggers the next steps in the data processing flow by sending data to the API Gateway.
• API Gateway & AWS Lambda:
• API Gateway triggers Lambda functions when new data is available from S3.
• Lambda functions interface with the machine learning component.
• Machine Learning Model:
• Processes data received from the Lambda function.
• Performs analysis, extracts resume data, and returns the processed results.
Architectural Patterns and Design Principles
• The project cleanly separates concerns between the frontend and backend.
• Uses a serverless approach to integrate external services like file storage (S3) and event-driven compute (Lambda) for ML processing.
• There’s an emphasis on modularity: each component (upload management, data processing, ML model) functions independently yet the system coalesces through clearly defined APIs and event flows.
Relevant Technologies and Frameworks
• Frontend: Next.js, React (with components like InfoCard, Navbar, etc.), Tailwind CSS.
• Backend: FastAPI (despite the README mentioning both Python and Node/Express; the actual file hints FastAPI in backend/main.py – verify if it’s a mix or if part of documentation may be legacy or complementary).
• AWS Components: S3 for storage, API Gateway, Lambda for serverless processing, ECS Fargate for containerized deployment.
• Machine Learning: The pretrained ML model found under the "model" directory, supported by a Jupyter Notebook for training/preliminary investigation.
Essential Elements
• Frontend → Backend (file upload and data requests).
• Backend ↔ AWS S3 (file storage and retrieval triggers).
• AWS S3 → API Gateway → Lambda (event-driven processing).
• Lambda → Machine Learning Model (data processing) and then results flow back, either directly to Lambda and then through the backend or directly into S3 for query.