3 ms·
Still a bad bet. Their moat is deeper now than it was in 2022. The engineers you need to poach are all paid well over 1+ million USD per year now. The number of
by Der_Einzige 2y ago
Still a bad bet. Their moat is deeper now than it was in 2022. The engineers you need to poach are all paid well over 1+ million USD per year now. The number of people capable of writing quality CUDA code to optimize transformer language models world-wide is likely less than 10000, and I'm being very generous. Nvidia holds a significant portion of that group, and some of the others you'll never find in the market at all since they hide behind discord profile pictures and mental illness.
- bjornsing 2y agoOnly reason that is hard is that CUDA is general purpose. You don’t need a graphics + GPGPU platform to run transformers. Nvidia is eating its own moat.