-
Welcoming start menu with a help button on left bottom.
-
User gets the instruction and puts the prompt into the user input if you want to make your own scenario or you can AI generate one.
-
This is the Gemini ai verdict after the user submits their defensive arguement for the made up scenario.
-
-
Results page where it displays what score you got each round and what is your overall score.
Inspiration
Ever watched a suspenseful court case movie? We wanted to recreate that intense feel by putting the user in their shoes in defending their client. As debaters we know it can be intimidating to speak up, especially for people from underrepresented. We believe that it is a valuable skill to have to be able argue and fight for what's right. So, we created Objection! It is an AI- powered debate arena where users can practice their debate skills and get meaningful feedback on your debate by AI.
What it does
We created a website that integrates AI as the judge to create an immersive feeling of being a lawyer taking you through different cases that are based on real life problems faced by underrepresented groups, enhancing your knowledge on critical thinking and debate. The program starts with user being given an option to choose an AI generated case or create a custom case. The user then has 2 minutes to come up with their defensive argument. After submission Gemini judge gives its verdict with key points on improvement and after completing 3 cases, you receive an overall score.
How we built it
We used React for the frontend, Node.js to connect the Gemini API in the backend for the case generation and verdict on how the user did, normal CSS for styling, and JavaScript for functionality. We also got help from chatGPT-5 and online documentation (for Gemini API) to help tie things together and clarify coding concepts.
Challenges we ran into
Our team started the hackathon with a busy day. One member had a chess tournament, while others had robotics. By the time we finally started we ran out of a lot of time for the Hackathon. Also, a big issue that lingered during the whole development process was the implementation of Gemini AI into our project. But in efforts of adding this we ran into many problems such as the API Key not working and Gemini not giving us the responses that we wished to see. We fixed this problem by finding documentation and guides on YouTube that helped us debug the issue.
Accomplishments that we're proud of
After this development, our team can easily say the thing we are most proud of is the implementation of Gemini into our Idea of debate. We definitely think we came up with a unique idea that encompasses the fun of debate with AI.
What we learned
We learned how to implement and use API, fetch, get information from API's especially AI generated responses that changes based on the user input! ** Then implementing it into our application!**
What's next for OBJECTION!
If presented with more time, we would continue to enhance the user interactions with the AI itself to be more conversational putting you directly in the courtroom itself giving a more intense and realistic experience. Also, we're planning to present to our school's debate team and hopefully others as a resource to practice for debate.
Log in or sign up for Devpost to join the conversation.