What to ExpectAs a member of the Dojo 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 massively parallel Dojo accelerators.  The ideal candidate is an initiative-taking,
proactive engineer to ensure Dojo software quality is high and user experience
is seamless.What You’ll DoDeveloping Python libraries to run, monitor, measure, and
troubleshoot deep learning workflows running on DojoThe role will be programming approximately 50% of the
time in Python. The remaining time will be spent debugging, experimenting, and
investigatingWriting tests to guarantee correctness at every level of
the Dojo stack, from high-level pytorch integration, through every stage of the
ML compiler stack, down to our custom hardwareEnsuring that neural networks of interest to our users
function correctly on Dojo at the expected high performance  Writing tools to run, monitor, measure, and troubleshoot
deep learning workflows running on DojoWriting bench-marking and reporting toolsSupporting in-house users; triaging errors and
performance bottlenecks, driving issues to root cause, providing workarounds
and fixesAnticipating likely use cases; solving problems before
our users hit them

What You’ll BringDegree in Engineering, Computer Science, or equivalent in experience and evidence of exceptional ability in related
fields with practical software engineering experienceStrong proficiency with Python and comfortable with C++Highly familiar with Linux administration and internals

PALO ALTO, California

Full time