CounselMate is an AI-powered career guidance platform designed to help students explore and discover suitable career paths based on their unique skills, interests, and educational backgrounds.
Developed as a major college project by
Nandini Goyal, Rishika Raj, and Vaishnavi Jadon
-
Natural Language Processing (NLP)
Uses Hugging Face’s Zephyr-7B-beta model for generating responses.
Extracts career interests and skills from user input using regex patterns. -
Personalization
Stores user preferences (skills, interests, education background) in MongoDB.
Dynamically recommends careers tailored to individual users. -
Conversation Flow
Follows a structured SYSTEM_PROMPT to simulate a professional career counselor.
Maintains conversational context with a 5-message history window.
Before running the project locally, make sure you have the following installed:
- Python 3.7+
- Node.js (v14+ recommended)
- pip (Python package manager)
git clone https://github.com/rishika2108/counselmate.git
cd counselmate- Log in to MongoDB Atlas
- Create a new cluster
- Select Python as the driver
- Copy the connection URI and paste it inside:
backend/config.py:
MONGO_URI = "your-mongo-uri"- Log in at huggingface.co
- Go to Settings → Access Tokens
- Generate a new token
- Paste the token in:
backend/config.py:
HF_TOKEN = "your-token"cd backend
pip install -r requirements.txt
python admin_creation.py
python data_seeder.py
python app.pyIn a new terminal:
cd counselmate-frontend
npm install
npm startThen visit: http://localhost:3000 to start using CounselMate!
Built with passion by:
- Nandini Goyal
- Rishika Raj
- Vaishnavi Jadon
This project is developed for educational and demonstration purposes only. It may be freely modified or extended for personal, academic, or non-commercial use. Commercial use requires permission from the authors.