ML Compiler Engineer, Compute

As a member of Waymo's Compute Team, your role is crucial in delivering the compute platform that powers the autonomous vehicle's software stack. You will be involved in architecting and creating high-performance custom silicon, developing system-level compute architectures, and collaborating with teammates to optimize hardware and software for maximum performance. This is an exciting opportunity to work on one of the world's highest performance automotive compute platforms. **Key Responsibilities:** - Maximize performance of neural networks by enhancing and extending the production-grade compiler. - Collaborate with hardware architects and model developers to understand the unique neural network inference platform and neural networks. - Implement compiler support for novel features of the high-performance architecture. **Qualifications Required:** - Bachelor's degree in Computer Science/Electrical Engineering or equivalent practical experience with at least 3 years of industry experience, OR - Master's degree in Computer Science/Electrical Engineering with at least 1 year of industry experience, OR - PhD Degree in Computer Science/Electrical Engineering or equivalent years of experience. - At least 1 year of industry and/or academic experience with compilers and parallel computing. - At least 1 year of industry and/or academic experience working with ML inference or linear algebra computations. - Proficiency in C++ programming. **Preferred Qualifications:** - Experience in Python programming. - Knowledge of compilers for neural networks. - Understanding of computer architectures used for neural network inference and neural network performance characteristics. - Familiarity with the principles behind popular machine learning and neural network algorithms and applications. Please note that the salary range for this position is INR 3,400,000 to 4,110,000. (Note: The above job description is based on the provided details and may not include any additional company information mentioned in the original JD.),