← All projects
Project record01 mention
Originally shared on Hacker News

leafle

A book-discovery app that recommends new books based on a deeper understanding of what the reader likes than typical 'people who read X also read Y' suggestions, and surfaces titles that can be borrowed right away from local libraries. It works with about 6,500 public libraries around the world and uses LLMs to reverse-engineer the reader's tastes.

Last shared For Mostly for personal use by the author
01 / The problem

What it addresses

Typical book recommendations ('people who read X also read Y') are too shallow and don't account for what the reader actually likes or what is immediately borrowable at their local library.

02 / Source history

Shared over time

Each entry reflects what was shared on that date. Details may have changed since then.

  1. leafle

    A book-discovery app that recommends new books based on a deeper understanding of what the reader likes than typical 'people who read X also read Y' suggestions, and surfaces titles that can be borrowed right away from local libraries. It works with about 6,500 public libraries around the world and uses LLMs to reverse-engineer the reader's tastes.

    Problem: Typical book recommendations ('people who read X also read Y') are too shallow and don't account for what the reader actually likes or what is immediately borrowable at their local library.

    Audience: Mostly for personal use by the author

    Posted by @kidnoodle · Original comment ↗

    “working on http://leafle.org my app (mostly for me) to find new books to read that I wouldn't have thought of (mostly recently Cosmicomics) based on a deeper understanding of what I like in books than people who read X also read Y, that I can actually borrow right away from my local library. Works on ~6,500 public libraries around the world.”