Software Engineer, Compiler, Scheduling and Optimization, Silicon

Google is seeking a Software Engineer to join their EdgeTPU compiler team, focusing on core optimization and scheduling algorithms for their next-generation compiler framework. This role is critical in delivering optimal AI/ML performance for Google's silicon technology to end users.

The position requires expertise in compiler development, particularly with Multi-Level Intermediate Representation (MLIR) and optimization algorithms. You'll be working on production-quality compilation and optimization of workloads, making a direct impact on hardware/software co-optimization for AI/ML use cases in current and future Google devices, including the latest Generative AI models.

As part of Google's mission to organize the world's information and make it universally accessible, you'll be working at the intersection of AI, Software, and Hardware to create groundbreaking technological experiences. The role offers competitive compensation ($141,000-$202,000) plus bonus, equity, and comprehensive benefits.

Key responsibilities include improving compiler quality and performance, developing parallelization and scheduling algorithms, optimizing Machine Learning workloads for EdgeTPU, and collaborating with architects on next-generation EdgeTPU architectures. You'll also work closely with product managers and researchers to identify emerging Machine Learning trends and future use cases.

The ideal candidate should have at least 2 years of experience in software development, particularly with C++ or Python, strong knowledge of data structures and algorithms, and experience with compiler optimizations and parallelization. Advanced degrees in Computer Science or related fields are preferred, along with experience in MLIR, LLVM, and machine learning architectures.

This role offers the opportunity to work on cutting-edge technology at one of the world's leading tech companies, with locations in Mountain View, CA or Bellevue, WA. You'll be part of a team that's pushing the boundaries of AI/ML performance and helping shape the future of Google's hardware capabilities.