It occurs to me that we can use LLMs to give better recommendations.

I want this for books.

The goodreads recommendations are crap.

The ton of “new hot books” newsletters I subscribe to are crap.

I want a daily email that makes highly specific recommendations made by an LLM based on:

  1. The books I am reading and have just read.
  2. The authors of those books.
  3. The genres of those books.
  4. The general areas of interest/study of those books.
  5. The specific details in my background.

I typically read on topics that interest me right now and I typically read a cluster of books on a topic.

I want to know about more/better books in the cluster.

I typically do this manually with an LLM and a ton of google/amazon/goodread searching, but it can be automated and made better.

Why stop at books?

I want LLM Recommended:

  • podcasts and podcast episodes
  • movies and tv shows
  • bands, albums, songs.
  • posts on socials
  • websites.

Suck up the consumption exhaust and give me better LLM-based recommendations.

“Consumption Exhaust” sounds cool. And accurate.

I could probably hack this together for books as a cron-bot in Python that uses the Goodreads API and OpenAI API and some carefully tuned prompts.

I may do this.