Book Finder

• 

  App image

Inspiration

As an avid reader and tech enthusiast, I noticed how overwhelming it can be to choose the right book when you're not sure what you're in the mood for. Recommendation platforms often feel impersonal, overloaded with irrelevant titles, or lack mood-based discovery. I wanted to create something smarter, an AI-powered book recommendation engine that genuinely understands the reader's intent, whether it's genre-based or emotion-driven.

What it does

Book Finder is a smart web app that recommends books based on a user's chosen genre or current reading mood. It allows users to pick a genre like Fantasy, Mystery, or Non-fiction, or select how they're feeling, Adventurous, Reflective, Romantic, etc. Based on this input, the app generates a curated list of book suggestions in real time, giving users a quick, engaging, and personalized reading experience.

How we built it

• I built Book Finder using bolt.new, which made it easy to rapidly prototype and deploy the app. Here's a breakdown of the stack:

• Frontend: Built with React and Tailwind CSS for a clean, responsive UI.

• Backend: Uses a simple Node.js serverless function to handle recommendations.

• Book Recommendation Logic: Leveraged OpenAI's language model via API to process user input and generate high-quality book suggestions based on mood or genre.

• Hosting: Deployed directly through bolt.new for quick iteration and testing.

Challenges we ran into

• Balancing creativity and relevance: Fine-tuning the AI prompts to ensure the recommendations were both creative and aligned with the user's mood or genre took a lot of experimentation.

• Handling ambiguous input: Users might input vague moods or genres, and building logic to interpret those correctly was a challenge.

• Response time: Optimizing API calls for fast results while maintaining quality required some backend tweaking.

Accomplishments that we're proud of

• Successfully created an intuitive and responsive UI that works seamlessly across devices.

• Built a mood-based recommendation engine that goes beyond simple keyword matching.

• Users have praised the experience as "fun," "unexpectedly accurate," and "addictively helpful," which validates the vision behind the app.

What we learned

• Small details in prompt engineering can make a big difference in output quality.

• Building with bolt.new is incredibly fast and effective for deploying smart web apps with serverless functions.

• There's a real user need for mood-based recommendation tools—not just in books, but potentially for movies, music, and more.

What's next for Book Finder

• User Accounts & Personalization: Let users save preferences, rate books, and get better suggestions over time.
• Integration with external APIs like Google Books to show covers, summaries, and purchase links.
• Mobile-first experience: Launch a PWA (Progressive Web App) version.
• Social Sharing: Allow users to share their book lists or moods with friends.
• Analytics & Feedback Loop: Use feedback to train a more customized recommendation model.

Built With

• googlebooksapi
• react
• tailwinds