[pirates]

For some reason everyone that says things like this never follow up with anything concrete, don’t share prompts or snippets, etc.

[ako]

This project is completely built using claude code: https://github.com/ako/backing-tracks Most of the features take less than 30 minutes.

[shimman]

This isn't that impressive when there are mountains of training data dealing with exactly this... how about something truly unique and not something already available to the masses in hundreds of different forms?

Like cool, you killed boiled a few gallons of the ocean but are you really impressed that you made a basic music app that is extremely limited?

[jdiez17]

This looks pretty neat, thanks!

[teaearlgraycold]

I’ve seen first hand people talk big about how they used LLMs on a project and it’s clear they’ve only done the first 80%. Yeah they’re good tools. But they also enable laziness.

[Robdel12]

https://www.npmjs.com/package/@vizzly-testing/honeydiff

I worked for Percy for 4 years. We were “stuck” with imagemagik to do diffing (I’m sure they still might). I was able to build my own differ with Claude/LLM help.

That special enough for you? Or?

[teaearlgraycold]

I looked at the source. It seems most of the code is not included in the GitHub repo, which itself contains a bit of JS glue. The .tgz uploaded to npm has various prebuilt binaries. Can I take a look at the rust code?

I'm not trying to imply LLMs aren't useful. I just want more info from GP so that I can evaluate their claims.