I would use RSS to aggregate the different sources I care about. You can use the unnoficial HN RSS feed to stay on-top of certain LLM keywords, and try to scrap

I would use RSS to aggregate the different sources I care about. You can use the unnoficial HN RSS feed to stay on-top of certain LLM keywords, and try to scrape together different reliable source websites that have Atom/RSS feeds. The end-result is a chronological feed of syndicated data, stored in one OPML file and portable across different apps. Your mileage may vary, but this works great for me.