What to ExpectAs a member of the Dojo compiler team, you will be responsible
for enabling Tesla’s neural networks to train efficiently on our upcoming
in-house custom-silicon supercomputer systems. Join a small team of experienced
developers in automating the compilation of PyTorch-derived neural network
graphs into programs that run on Tesla’s custom FSD computer. The ideal
candidate is an experienced compiler engineer comfortable working rapidly in a
small-team environment. 

What You’ll DoTake ownership of a few areas of the compiler (flexible,
based on skills/interests/needs) Develop algorithms to improve performance and reduce
compiler overhead Debug functional and performance issues on massively
parallel systems Collaborate with Dojo HW team to understand current HW
architecture and propose future improvements Work with Autopilot SW team to assure smooth transition
of training from GPU to Dojo 

What You’ll BringDegree in Engineering, Computer Science, or equivalent in experience and evidence of exceptional abilityPrior industry or research experience within compiler developmentComfortable with C++ and assembly code

PALO ALTO, California

Full time