AI Compiler Engineer, Autonomy

Tesla

  • Full Time

The Role: 

As a member of the Autonomy group, you will have the opportunity to apply your technical skills to a variety of system components & foundational code targeting higher performance of Autopilot and our Humanoid robot. You will be responsible for the internal working of the AI compiler running Neural Networks in millions of Tesla cars. Having experience or familiarity with Computer Vision, Machine Learning & related software concepts is a plus. You’ll work with many of the same great engineers who delivered Tesla’s custom FSD computer as well as Machine Learning scientists.  The ideal candidate is an experienced compiler developer comfortable with working rapidly in a small-team environment.  

 

Responsibilities: 

  • Take ownership of a few areas of the compiler (flexible, based on skills/interests/needs).
  • Closely collaborate with AI team to guide them on the design and the development of Neural Networks into production. 
  • Collaborate with HW team to understand current HW architecture and propose future improvements. 
  • Develop algorithms to improve performance and reduce compiler overhead. 
  • Debug functional and performance issues on massively-parallel systems. 
  • Work on architecture-specific neural network optimization algorithms for high performance computing.

 

Requirements: 

  • BS in relevant field (Computer Science, Computer Engineering, etc.) 
  • Prior industry experience as a compiler developer 
  • Comfortable with C++ and assembly code 
  • Capable of delivering results with minimal oversight 

Nice to Have: 

  • MS in related field (Computer Science, Computer Engineering, etc.) 
  • Previous professional experience with Python 
  • LLVM code familiarity 
  • Performance analysis experience 
  • Experience coding parallel programs