Google is seeking a Software Engineer III to join their ML Compiler Infrastructure team. This role focuses on device code generation for GPU, specifically generating efficient instructions to execute ML models and maintaining infrastructure for compiler engineers. The position is critical for enabling high-performance inference and training of first- and third-party models on GPU across Alphabet.
The ideal candidate will have strong C++ programming skills and experience with large-scale infrastructure or distributed systems. Knowledge of GPU optimization and compiler fundamentals is highly valued. The role involves working with cutting-edge ML technologies and requires both technical depth in systems programming and the ability to collaborate effectively with compiler engineers and partners.
As part of Google Cloud, you'll be contributing to technology that helps organizations digitally transform their businesses. The team's work directly impacts Google's ability to serve billions of users efficiently through optimized ML model execution. This is an opportunity to work on fundamental infrastructure that powers Google's ML capabilities while being part of a team that pushes the boundaries of performance optimization.
The position offers exposure to complex technical challenges in ML infrastructure, compiler optimization, and GPU computing. You'll be working in an environment that values innovation and technical excellence, with opportunities to make significant contributions to Google's ML infrastructure. The role combines hands-on coding with architectural decision-making, making it ideal for engineers who enjoy both detailed technical work and broader system design.