NeonBeam

NeonBeam: Because Even Dinner Deserves a Spotlight!

Comment

Project Story: NeonBeam

Inspiration

NeonBeam was inspired by the common struggle of decision paralysis—those moments when you can’t figure out what to eat, what movie to watch, or even what to wear. As a team, we realized how often we overthink small decisions and thought, Why not make this process fun? Our goal was to create a tool that turns indecision into an engaging and entertaining experience, offering a touch of humor to lighten the burden of choice.

What it does

NeonBeam helps people make decisions in a playful and interactive way. Here’s how it works:

  1. Users enter a question, such as “What should I eat for dinner?”
  2. NeonBeam uses an AI-powered Large Language Model (LLM) to generate a list of potential options.
  3. A dynamic word cloud visualizes the suggestions, offering a fun way to explore possibilities.
  4. Simultaneously, an interactive option selector begins cycling through suggestions.
  5. When the user presses the stop button, the cycling stops on a final, randomly chosen solution.

It’s like spinning a digital decision wheel but smarter and more visually engaging!

How we built it

  • Frontend: Built using HTML, CSS, and JavaScript to ensure a clean and user-friendly interface.
  • Backend: Powered by a fine-tuned Gemini-1.5-Flash model, which generates creative and context-aware options tailored to the user’s input.
  • Visualization: Leveraged a word cloud library to dynamically display AI-generated suggestions in real time, enhancing the visual appeal of the platform.
  • Server: Deployed on Vercel, utilizing its serverless infrastructure for fast and reliable performance.
  • Gamified Selector: Designed a JavaScript-based mechanism to rotate through options, adding suspense and fun to the decision-making process.

Challenges we ran into

  1. Optimizing the Word Cloud: Ensuring smooth real-time updates while maintaining a visually appealing layout was tricky.
  2. Time Management: With limited hackathon hours, balancing ambition and feasibility was a constant challenge.
  3. Integration: Combining the backend AI with the frontend visualization required meticulous debugging to ensure everything worked seamlessly.

Accomplishments that we're proud of

  • Successfully built a gamified decision-making platform that is both functional and entertaining.
  • Developed a project that’s lighthearted and approachable while showcasing practical applications of AI technology.

What we learned

  • Model Fine-Tuning: Developed skills in fine-tuning the Gemini-1.5-Flash model to create more tailored and contextually relevant outputs for user queries.
  • Gamification Design: Learned how to craft engaging, user-centric features that combine functionality with entertainment to create a fun and interactive experience.
  • Serverless Deployment: Gained experience in using Vercel’s serverless infrastructure for seamless and efficient deployment.
  • Team Collaboration: Strengthened our ability to coordinate effectively under tight deadlines, balancing creative problem-solving with technical implementation.

What’s next for NeonBeam

  • Enhanced Customization: Allow users to set parameters like the number of options or specific categories for suggestions.
  • Mobile Optimization: Build a mobile-friendly version to make NeonBeam accessible anywhere.
  • Expanded Use Cases: Integrate APIs for local restaurant recommendations, movie listings, and more to provide highly relevant suggestions.
  • Social Sharing: Add a feature for users to share their final choices with friends for added fun.

Built With

Share this project:

Updates