1 open role mentioning MXFP8
Thinking Machines Lab
Thinking Machines is seeking an infrastructure research engineer to design and build the core systems that enable efficient large-scale model training, with a particular focus on numerics. This role involves improving the numerical foundations of their distributed training stack, from precision formats and kernel optimizations to communication frameworks, ensuring the stability, scalability, and speed of training trillion-parameter models. The ideal candidate will excel at the intersection of research and systems engineering, possessing a strong understanding of both optimization mathematics and the practicalities of distributed compute.
$350k - $475k