Give your user a game while they wait for the LLM to return a result.

That is awesome!

It would help if you had a video or gif. I did command + shift + 5 on a mac and recorded. Asked the coding agent to speed it up, edit out some parts between timestamps, and put it in the docs. It did all that. I started an issue anywhere on GitHub, clicked the upload button, uploaded it, grabbed the URL, the coding agent did the rest. It was extremely easy. (Not too late!)

I'm trying to turn my personal webpage into Breakout.

If you are looking for some high quality simple JavaScript games to incorporate have a look at Jake Gordon's Games [0]

[0] https://jakesgordon.com/games/

Could also be a quiz or a mini lecture. It knows all about the context you are in and the problem you are trying to solve, so might as well learn something.

Like many others starting out, I built agents with personalities, multiple roles at each level etc. A great way to burn tokens but not effective at all in producing better outcomes.

However, one great side effect was the learning I got from seeing the Marc Andreessen agent hash it out with the Ben Horowitz agent etc. That felt like the future of online courses.

Now's your chance to pull a Namco and file an equivalent patent for "LLM is working" minigames!

https://en.wikipedia.org/wiki/Loading_screen#Minigames

I was really hoping for a gif or two!

we've reached another level of dopamine-hit chasing....

where are they hosted so i can play?