> The code uses browser automation because creating a RegEx pattern for all weird-as-heck "company" names is hard.

That doesn't explain why web scraping is used instead of using the GitHub API, which is more kosher.

With the API, query all Stargazers, then query the user profiles of those users.

For unauthenticated requests, this will only work if n < 60, but for authenticated requests, the sky is the limit.

I have an old GitHub repo that essentially implements this and I believe it still works: https://github.com/minimaxir/get-profile-data-of-repo-starga...

It only needs a tweak to support the company field, and I might do just that if there is enough demand.