Overview DebateAI is a live AI system that evaluates arguments in real time. It was inspired by how debates are often judged based on opinion rather than actual reasoning, especially in school settings. The goal was to make debates more fair and help people clearly understand why one argument is stronger than another.
How It Works The project is built as a full-stack web application. The frontend uses React to create a clean and interactive interface, while the backend handles logic and API calls. An AI API analyzes each message in the debate.
For every message, the system:
classifies it as a claim, evidence, or rebuttal assigns scores for logic, evidence, and clarity detects basic logical fallacies
These results are sent back and displayed in real time, updating scores as the debate progresses.
Scoring Model The scoring system is based on averaging the quality of arguments over time:
Final Score = ∑(Logic+Evidence+Clarity)/total messages
This ensures consistent evaluation instead of relying on a single moment.
What I Learned Through this project, I learned how to design a real-time system and connect AI outputs to a working application. I also learned the importance of UI design when presenting complex data, as well as how to manage time and focus on core features.
Challenges One challenge was making the system feel fast while processing AI responses. Another was ensuring the AI outputs were consistent and structured enough to use effectively. Balancing advanced features with reliability was also difficult. Additionally, polishing the interface to make it feel like a real product required more effort than expected.
Conclusion DebateAI turns debates into a more structured and understandable process. Instead of relying only on opinion, it provides clear reasoning behind each result and helps users improve their arguments.
Built With
- 18
- css
- gemini-3-flash-preview
- genai
- motion
- nginx
- node.js
- react
- recharts
- sdk
- tailwind
- typescript
- vite
Log in or sign up for Devpost to join the conversation.