---
description: Details the technology stack chosen for the CV Optimization Platform.
globs:
---
# Technology Stack

## Frontend
- **Framework:** React.js (using Next.js for server-side rendering and routing)
- **Styling:** Tailwind CSS
- **State Management:**  (Initially simple React state, consider Context API or Zustand if complexity grows)
- **UI Library:** (Consider a component library like Material UI or Ant Design if needed for more complex UI elements later)

## Backend
- **Language:** Node.js (with Express.js framework) or Python (with Flask/FastAPI - *Decision needed based on your preference and libraries for LLM interaction*)
- **API Framework:** Express.js (Node.js) or Flask/FastAPI (Python)
- **Document Processing Libraries:**
    - Python: `PyPDF2`, `python-docx` (if using Python backend)
    - Node.js: `pdf-parse`, `docx-parser` (if using Node.js backend)

## Database
- **Initial Stage:** File-based storage (for simplicity in MVP) - *Consider moving to MongoDB or PostgreSQL for user accounts and scalable data storage later.*
- **Future Scalable Database:** MongoDB (NoSQL) or PostgreSQL (Relational) - *To be decided based on data structure and scalability needs.*

## LLM Integration
- **Primary LLM Provider:** OpenAI API (GPT models) - for initial development and ease of use.
- **Potential Alternative:** Hugging Face Transformers - for exploring open-source models and potential cost optimization in the future.

## Deployment
- **Frontend Hosting:** Vercel (for Next.js applications)
- **Backend Hosting:** Render or Heroku (or AWS/Google Cloud/Azure for more control and scalability)