Google is seeking a Staff Software Engineer specializing in Machine Learning Compilers to join their EdgeTPU team. This role combines cutting-edge AI technology with hardware optimization, requiring expertise in compiler development and machine learning frameworks. The position offers an opportunity to work on next-generation technologies that impact billions of users worldwide.
The ideal candidate will have extensive experience in software development, particularly with compilers and machine learning systems. You'll be working with advanced ML frameworks like JAX and PyTorch, optimizing them for EdgeTPU deployment. The role involves close collaboration with hardware architects and ML researchers to bridge the gap between research innovations and practical implementations.
Key responsibilities include extending ML authoring frameworks, optimizing model performance, and improving compiler quality. You'll work at the intersection of hardware and software, designing interfaces and co-optimizations between CPU, GPU, and TPU. This position requires both technical depth in compiler development and the ability to collaborate across teams.
The compensation package is highly competitive, ranging from $189,000 to $284,000 base salary, plus bonus, equity, and comprehensive benefits. This role is based in Mountain View, CA, at Google's headquarters, where you'll be part of a team that's pushing the boundaries of AI and hardware acceleration.
Google offers a collaborative environment where you'll work with talented engineers and researchers, contributing to projects that have global impact. The company's commitment to innovation in AI and hardware makes this an exciting opportunity for someone passionate about compiler technology and machine learning optimization.
This role represents a unique opportunity to shape the future of AI hardware acceleration while working with state-of-the-art technology at one of the world's leading tech companies. You'll be at the forefront of developing solutions that make AI more efficient and accessible across Google's product ecosystem.