@on1zuma Done! 😊

@on1zuma That would definitely improve speed. Will still need to retrieve from backend for new articles though. Perhaps could implement a hybrid combination retrieving only after latest date for topics within the stored session. Will need to think more on it. Thanks for the suggestion 👍

@nikolay Hi @nikolay! Thanks heaps, really appreciate your support. Currently planning a "save article" feature for the next update, and your suggestion for deeper integrations is great - will definitely keep that in mind for future improvements.

@on1zuma Good to hear, I'm doing well. Thanks Wassim, much appreciated! The article and topic data, along with their embeddings, are stored in a db. I then use FAISS to semantically match articles to the topics by comparing the embeddings. Still needs some work to improve speed though.

@on1zuma Hi Wass, how's it going?

@theresanaiforit provides a free-to-read platform where you catch up on the news for your favourite topics in bite-size recaps. Each topic is curated with the assistance of AI and every recap can be expanded on for a deeper understanding of the story at hand.