I asked a few of my friends that are ML engineers this question and all of them said to run the LLMs in the cloud with their infrastructure because it was going to be way faster.
If you just want to tinker around I would look at @JSR_FDD's comment.
I finally kicked off my browser based game with the help of codex cli this year. It has been like having a junior developer that takes decent direction. There have been some missteps but "we" have made a lot more progress on it than I had the last three years. And it let's me spend my time optimizing the things I like doing.
I feel like most first intro classes in Computer Science is learn the coding language on your own. At first I was like why? Why don't they hold our hands while we do this. But since I have had some space to look back it really is a pretty good representation of our industry. You are going to need to learn new languages. So getting thrown in the deep end is a pretty good precursor for what work is going to look like.
I find that for most seniors, you should follow the documentation with check ins from current employees. If the documentation is obtuse then have the new hire correct it so that the next new hire having the same information at that point will understand what to do better.