Viberary is a search engine that recommends you books based not on genre or title, but vibe by performing semantic search across a set of learned embeddings on a dataset of books from Goodreads and their metadata.
It returns a list of book recommendations based on the vibe of the book that you put in. So you don’t put in “I want science fiction”, you’d but in “atmospheric, female lead, worldbuilding, funny” as a prompt, and get back a list of books. This project came out of experiences I had where recommendations for movies, TV, and music have fairly been good, but book recommendations are always a problem.
Here is my (@cogdog) attempt and the results have a good vibe!
This is one among many items I will regularly tag in Pinboard as oegconnect, and automatically post tagged as #OEGConnect to Mastodon. Do you know of something else we should share like this? Just reply below and we will check it out.
This project is a lot of fun, but conclusively proves to me what I’ve known all along about myself: reaching MLE (machine learning enlightenment) is the process of working through modeling, engineering, and UI concerns, and connecting everything together - the system in production is the reward. And, like any production-grade system, machine learning is not magic. Even if the data outputs are not deterministic, it takes thoughtful engineering and design choices to build such a system, something that I think gets overlooked these days in the ML community. I hope with this write-up to not only remind myself of what I did, but outline what it takes to build a production machine learning application, even a small one with a pre-trained model, and hope that people scope their efforts accordingly.
I can’t understand all the technical details but I respect this level of transparency.
It is the opposite of the AI heavyweights who are opaque and even make poor use of the word “open” in the company name.