Project Info

TrivAI

This project did not submit a demo video on Devpost.

Inspiration

TrivAI was born from the desire to revolutionize the traditional trivia game experience. My girlfriend and I like trivia and we noticed that conventional trivia games often lack personalization, asking questions about topics we don't care about. By leveraging AI technology, TrivAI seizes an opportunity to create dynamic, engaging, and customizable trivia experiences that adapt to players' interests, helping us learn more about our favorite topics!

What it does

TrivAI is an interactive multiplayer trivia game that uses AI to generate unique questions based on player-chosen topics. Key features include: Real-time multiplayer gameplay using Pusher for live updates Custom topic selection AI-powered question generation using gpt-4o Dynamic scoring system Instant answer validation using AI Fun, playful UI with bouncing shapes and smooth transitions

How we built it

We built TrivAI using a modern tech stack: Frontend: Next.js with TypeScript for type safety and better developer experience Styling: Tailwind CSS for responsive design and custom animations Real-time Communications: Pusher for websocket connections and live game updates AI Integration: OpenAI's gpt-3.5-turbo for question generation and answer validation State Management: Server-side global state for managing game sessions and player data

Challenges we ran into

Answer Validation: Creating a fair and accurate system for validating player answers was complex. We solved this by using GPT-3.5 Turbo to compare answers contextually rather than exact matching. Real-time Synchronization: Coordinating game state across multiple players presented challenges. We implemented a robust Pusher-based system to ensure all players stay synchronized. Question Quality: Ensuring AI-generated questions were both challenging and accurate required careful prompt engineering and validation.

Accomplishments we're proud of

Created a seamless multiplayer experience with real-time updates Implemented an animated UI that gives the game a fun vibe Successfully integrated AI for both question generation and answer validation Built a scalable architecture that can handle multiple concurrent game sessions

What we learned

Websocket implementation with Pusher Prompt engineering for AI interactions State management in a real-time multiplayer context Complex animation implementations with CSS TypeScript best practices in a Next.js environment

What's next

Custom Game Modes: Implement different game modes like time trials, tournament-style competitions, and Jeopardy style betting Difficulty adjustment based on player performance Friend lists and public lobbies Achievement system A dashboard for players to track their performance and learning progress over time

Analysis

Compare with all teams

View

Metric

Kavel Rao
9

Figures cover GitHub contributors during the hackathon window. A co-authored commit counts in full for each author, so per-member totals add up to more than the whole-team figures.

Technology

Found in codeClaimed only

CSSIn code
JavaScriptIn code
Next.jsIn code
OpenAIIn code
ReactIn code
Tailwind CSSIn code
TypeScriptIn code
Vercel AI SDKIn code

8 of 8 appear in the indexed code.

AI coding agents

No AI coding agent signals were found in this repository.

Detected from committed agent config files and commit authorship. Absence of a signal is not proof an agent was unused.

Codebase size

Source size

46 KB

Source files

Counts recognized source files only; vendored directories, binaries and lockfiles are excluded, so this is smaller than the repository on disk.