My current project is TrendBowl (https://www.trendbowl.app) which summarises trending entities across multiple sources on the web, including Mastodon and Wikipedia. On a 24 hour basis you can see which terms are relevant in different languages and sites.
It is implemented in Go, Python and JS, with support from Postgres, TimescaleDB and Kubernetes.
Why no Twitter/X? I believe they have an API that you can use, but they've locked down scraping pretty hard. The API should be cheap however, if you're only scraping trending once per day.
Originally I had Twitter support, but they changed the API around and it ended up not working any more, see the project blog for more info: https://blog.trendbowl.app/
The Pro subscription to the API now seems to be $5000 per month!
It is implemented in Go, Python and JS, with support from Postgres, TimescaleDB and Kubernetes.