BarraCUDA Open-source CUDA compiler targeting AMD GPUs

Re:BarraCUDA Open-source CUDA compiler targeting AMD GPUs by ZaneHam on Wednesday February 18, 2026 @03:46AM (#47058779)

Hello, I’m the project author. I don’t think In any of this and some of the criticisms I’ve received on this forum have people realised I’m not the original poster. I posted this on R/compilers and as of now that’s pretty much it. In terms of the comments. I use intellisense from time to time, I put my own humour into things and because that’s who I am. I’m allowed to do these things.

I’m self taught in this field. I was posting on R/compilers and shared this around with some friends who work within this space for genuine critique. I’ve been very upfront with people on where I use LLMs. It’s actually getting a bit “too much” with the overwhelming attention.

[ Parent | on HN ]

Re:BarraCUDA Open-source CUDA compiler targeting AMD GPUs by kmaitreys on Wednesday February 18, 2026 @04:30AM (#47059041)

I understand your position. If I were in your place where someone else posted my personal (?)/hobby project on a public forum where it got discussed more on the point if it was LLM generated or not rather than the more interesting technical bits, I would also be frustrated.

Regarding the writing style, it's unfortunate that LLMs have claimed a lot of writing styles from us. My personal opinion is to avoid using these AI-isms but I completely get that for people who wrote like that from the start, it's quite annoying that their own writing is now just labelled as LLM generated content.

[ Parent | on HN ]