Gemini.Orbit

Visualizing the Universe of Knowledge_

Comment

Inspiration

Learning complex topics like Quantum Physics or Neural Networks from flat text is hard. We wanted to transform the way people learn by turning abstract concepts into interactive 3D constellations. Our goal was to make knowledge explorable, visual, and beautiful breaking away from static pages into a dynamic universe of ideas.

What it does

Gemini.Orbit is a 3D knowledge graph generator powered by Google's Gemini AI.

  • Visualize: Enter any topic (e.g., "Black Holes"), and watch it spawn a dynamic 3D graph where nodes are concepts and edges are the hidden connections between them.
  • Interact: Rotate, zoom, and dive into the knowledge verse. Click on any node to get instant, AI-generated explanations for that specific concept.
  • Save: Build your personal library of knowledge graphs. Save your "Orbits" to our secure cloud database and revisit them anytime to continue your learning journey.

How we built it

We engineered a modern, high-performance full-stack application:

  • Frontend: Built with Next.js 14, TypeScript, and TailwindCSS for a sleek, responsive UI. We utilized react-force-graph and Three.js to render complex 3D visualizations directly in the browser.
  • AI Engine: Integrated Google Gemini Pro API to analyze prompts and generate structured graph data (nodes and relationships) in real-time.
  • Backend: Developed a robust Java Spring Boot REST API to handle business logic, data processing, and user management.
  • Database: Leveraged Neon (Serverless PostgreSQL) for scalable storage of user orbits and graph data.
  • Authentication: Implemented Clerk for secure, seamless user authentication and session management.
  • Cloud Deployment: Dockerized the Spring Boot backend on Koyeb and deployed the frontend on Vercel for global availability.

Challenges we ran into

The biggest hurdle was orchestrating a distributed full-stack architecture. Connecting a Vercel-hosted Next.js frontend with a Koyeb-hosted Java backend introduced complex CORS issues and secure communication challenges. We solved this by implementing strict security configurations and proxy handling. Additionally, prompt engineering for Gemini was crucial—we iterated extensively to ensure the AI always returned valid JSON structures that our 3D engine could parse without errors.

Accomplishments that we're proud of

  • Successfully deploying a production-grade Java + Next.js full-stack app completely for free using modern cloud platforms.
  • Creating a visual experience that feels premium and futuristic using advanced 3D rendering techniques.
  • The lightning-fast speed at which Gemini generates complex knowledge graphs from simple text prompts.

What's next for Gemini.Orbit

  • Real-time Collaboration: Letting multiple users explore and build the same knowledge graph together in the same session.
  • VR Support: Taking the experience to the next level by allowing users to explore these knowledge graphs in Virtual Reality.

Built With

Share this project:

Updates