lumina Reads

Inspiration

Too much to read, not enough signal, especially when you’re growing skills, not just scrolling. Lumina Reads is for people who want clear, useful takeaways from real books and PDFs without drowning in noise.

What it does

You add books (including PDFs), read in the app, track progress, and get AI-backed summaries, insights, and chat tied to your library, so the material stays organized and actionable.

How we built it

A React + Vite front end with Firebase for auth and data, a Node API for library and AI flows, Gemini as LLM provider, and PDF text extraction plus reader UI for a smooth in-app reading experience.

Challenges we ran into

Reliable PDF handling (rendering, text, large files), keeping AI outputs grounded in what you actually uploaded, and auth + API wiring so the client, backend, and Firestore stay in sync.

Accomplishments that we're proud of

An end-to-end reading loop: upload → read → progress → insights/chat, with a UI that feels cohesive on a dark, focused reader experience.

What we learned

Product shape beats model hype: the hard parts are trust (accurate, cite-able-feeling help), performance with real documents, and simple flows so reading stays the hero.

What's next for lumina Reads

Smarter highlighting and notes, better source grounding for answers, reading goals and streaks, and polish for sharing (export summaries, study packs) so the app fits school, exams, and deep reading alike.

Built With

• a-node-api-for-library-and-ai-flows
• gemini-as-llm-provider
• react-+-vite-front-end-with-firebase-for-auth-and-data