Inspiration
The inspiration came from observing how different AI personalities affect user engagement. We wanted to test the hypothesis that personality matters more than features in chat applications. The idea was to create distinct AI characters that users could interact with to see which personality type generates the most engagement and conversation.
What it does
Multi Personality Bot is a chat platform featuring 10 distinct AI personalities:
- Compliment Bot (💖) - Always positive
- Rude Bot (😠) - Brutally honest
- Sarcastic Bot (😏) - British wit
- Motivational Bot (🚀) - High-energy
- Philosophical Bot (🤔) - Deep thinker
- Chaotic Bot (🎭) - Unpredictable
- Disagree Bot (⏳) - Always disagrees
- Argue Bot (⚔️) - Loves debates
- Movie Quotes Bot (🎬) - Film references
- Emotional Bot (🥺) - Highly empathetic
Users can chat with different bots to test which personality generates the most engagement. The platform tracks user interactions and provides real-time AI responses using Google Gemini API.
How we built it
Built using Python Flask for the backend, SQLite for data storage, and Google Gemini API for AI responses. The frontend uses HTML, CSS, Bootstrap 5, and JavaScript with a terminal/code theme. Each personality has unique prompts, response patterns, and engagement tracking. The system includes real-time chat functionality with Socket.IO.
Challenges we ran into
- Integrating Google Gemini API with proper error handling
- Creating distinct personality behaviors that feel authentic
- Managing real-time chat functionality across different personalities
- Ensuring consistent emoji rendering across different browsers
- Balancing personality traits to avoid offensive content while maintaining character authenticity
Accomplishments that we're proud of
- Successfully implemented 10 distinct AI personalities with unique behaviors
- Created a hypothesis testing framework for engagement metrics
- Built a responsive terminal-themed UI that works across devices
- Integrated real-time chat with proper error handling
- Developed a data collection system for personality preference analysis
What we learned
- Personality design in AI requires careful balance of traits and responses
- User engagement varies significantly based on AI personality type
- Real-time chat systems need robust error handling
- Emoji rendering can be inconsistent across different browsers and devices
- Hypothesis testing in AI applications requires careful data collection and analysis
What's next for Multi Personality Bot
This can be very helpful for review analysis and can be integrated in games and other platforms to keep users engaged. We will try to incorporate sentiment analysis with scores and all. The system can be expanded to analyze user sentiment across different personality interactions, providing valuable insights for user engagement optimization.
Built With
- bootstrap-5
- css
- flask
- font-awesome
- google-gemini-api
- html
- javascript
- primer-css
- python
- socket.io
- sqlite

Log in or sign up for Devpost to join the conversation.