🧠 CareerLens: Unsloth-PEFT-Based Multilingual Summarization and Dashboard Integration for Career Counseling Meets

CareerLens is an AI-powered platform designed to simplify and enhance the process of career counseling. By analyzing meeting transcripts (live or uploaded), CareerLens provides:

• ✅ Well-structured summaries
• 📌 Action items
• 💡 Insights
• 👥 Speaker recognition

It includes a powerful Attendee Bot that can automatically join meetings and a sleek frontend to manage, view, and interact with meeting summaries.

---

👥 Team

Name	LinkedIn / GitHub Profile
Dr. (Mrs.) Nupur Giri	LinkedIn
Piyush K. Chugeja	LinkedIn / GitHub
Manraj Singh Virdi	LinkedIn / GitHub
Sakshi Kirmathe	LinkedIn / GitHub
Deven Bhagtani	LinkedIn / GitHub

---

Project Video

🗂️ Project Structure

Module	Description
frontend	React-based interface built using Vite for user interaction
backend	FastAPI server to handle API requests and model inference
attendee-bot	Dockerized bot that joins meetings, records audio, and sends for processing

---

🔧 Prerequisites

Make sure you have the following installed:

• Node.js (v16 or later)
• Python (v3.9 or later)
• pip (Python package manager)
• Virtualenv (recommended)
• Docker (required for Attendee Bot)
• HuggingFace account (for HF token)

---

🚀 Getting Started

📥 Clone the Repository

git clone https://github.com/VESIT-CMPN-Projects/2024-25-BE03.git
cd 2024-25-BE03

---

🔌 Backend Setup (FastAPI)

cd Semester\ 8/backend
python -m venv venv
source venv/bin/activate  # For Windows: venv\Scripts\activate

Install all requirements.

Create a .env file and add:

ATTENDEE_API_URL=http://localhost:8000/api/v1/bots
ATTENDEE_API_TOKEN=<Your attendee API token>
AWS_REGION=<Your S3 bucket region>
AWS_ACCESS_KEY_ID=<Your AWS Access Key ID>
AWS_SECRET_ACCESS_KEY=<Your AWS Secret Access key>
HF_TOKEN=<Your HuggingFace token>

Start the server:

python main.py

The backend is live at http://localhost:5000

---

🤖 Attendee Bot Setup (Docker)

Clone the Attendee Bot from Attendee

Attendee bot runs at http://localhost:8000

---

💻 Frontend Setup (React + Vite)

cd Semester\ 8/frontend
npm install

Create a .env file:

VITE_TOGETHER_AI_API_KEY=<Your Together API Key>
VITE_API_URL=http://127.0.0.1:5000/api

Run the frontend:

npm run dev

Frontend is live at http://localhost:5173

---

✅ How to Run

Ensure all 3 modules are running:

Component	URL
Backend	http://localhost:5000
Attendee Bot	http://localhost:8000
Frontend	http://localhost:5173

Then open your browser at http://localhost:5173.

---

✨ Key Features

• 🧑‍💼 Meeting Management: Schedule, join, and manage meetings.
• 📄 Transcript Analysis: Upload meeting transcripts or audio for automatic summarization.
• 📊 Insights Dashboard: Get summaries, action items, and key takeaways with speaker roles.
• 🌐 Multilingual Support: Works with English, Hindi, and Marathi (with potential for more).
• 📁 Downloadable Reports (coming soon): PDF summaries and insights.

---

🛠️ Troubleshooting

• Check .env files for missing tokens or API URLs.
• Docker must be running for the Attendee Bot.
• If a port is already in use, update the port number in .env or in the terminal command.
• Use venv to isolate Python packages.

---

📄 License

This project is licensed under the MIT License. See the LICENSE file for details.

---

🤝 Contributing

Pull requests are welcome. For major changes, open an issue first to discuss your proposal.

---

📣 Acknowledgments

• 🤖 Unsloth for efficient model training
• 🤝 Hugging Face for models
• 🚀 Kaggle & Google Colab for compute support

---

Made with ❤️ for smarter, accessible, and inclusive career counseling.