A minimalist, symbolic Q&A interface powered by AI.
Built as a learning project to explore modern web development techniques including Next.js, PostgreSQL, Docker, and AI model integration.
- Symbolic Q&A interface with themed background
- Seamless fallback across multiple AI models
- Animated feedback for input errors and response delays
- User-friendly, mobile-responsive layout
- Personalized question history: view, search, delete
- Frontend: Next.js 15 (App Router) + TypeScript + TailwindCSS + Framer Motion
- Backend: PostgreSQL + Prisma ORM + Server Actions + REST API routes
- AI: OpenRouter API for multi-model integration
- DevOps:
- Docker for local PostgreSQL containerization
- Railway for production PostgreSQL hosting
- Vercel for full-stack Next.js deployment
- Built the core Q&A interface using Next.js App Router and TailwindCSS
- Integrated multiple AI models via OpenRouter with fallback support
- Explored:
- UI feedback through subtle animations (shake, typewriter, motion transitions)
- Prompt/response design patterns
- Symbolic and multilingual response handling
- Added PostgreSQL integration to support persistent question history
- Implemented:
- Search, pagination, deletion of history entries
- Backend logic via API routes and Server Actions
- Production database hosting on Railway
- Development database management using Docker Compose
- User authentication and saved sessions
- Personalized prompt suggestions based on history
- AI persona switching
- Admin dashboard for model logging and analytics
# clone the repo
git clone https://github.com/gracefy/answer-book-ai.git
# install dependencies
npm install
# Set environment variables in .env.local
OPENROUTER_API_KEY=your-api-key-here
DATABASE_URL=your-db-url-here
# start the dev server
npm run dev