A Gradio-based AI chatbot that acts as a virtual persona for a personal website.
This project is designed to answer questions about an individual's career and background by leveraging their LinkedIn profile and a personal summary.
A key feature is its self-evaluation mechanism, ensuring that all responses are high-quality and accurately reflect the individual's persona.

• Engages users in a professional, conversational manner, acting as a digital representative of an individual.

• Uses data from:
  • LinkedIn Profile (Profile.pdf)
  • Personal Summary (summary.txt)
• Provides informed and relevant answers.

• Uses an AI-powered evaluator to review and refine answers.
• If a response is subpar, it is regenerated based on constructive feedback before being sent to the user.

• The chatbot can:
  • Record contact information for interested users.
  • Log unanswered questions.
• Triggers push notifications via Pushover for real-time alerts.

The application is built around three core Python modules:

1. The chatbot receives a user's message.
2. A context-rich prompt is generated using the LinkedIn profile and personal summary.
3. The prompt is sent to the Gemini API for response generation.
4. The response is passed to a separate evaluator agent:
   • If approved → The response is sent to the user.
   • If rejected → A new response is generated using evaluator feedback.
5. During the process, the AI can trigger tools to:
   • Record data
   • Log unanswered questions
   • Send push notifications via Pushover

• Python
• Gradio – Interactive chat interface
• Gemini API – Response generation
• Pushover API – Real-time notifications

career-ego/
├── main.py         # Entry point for the chatbot app
├── prompts.py      # Builds prompts from LinkedIn + summary
├── tools.py        # External integrations (logging, notifications, etc.)
├── Profile.pdf     # LinkedIn profile data
├── summary.txt     # Personal career summary
└── README.md       # Project documentation