Software Engineer III, Cloud TPU Compiler
In this role, you will advance the PyTorch-in-Cloud offering by enabling PyTorch models to run efficiently on Cloud TPUs and GPUs and tuning for peak performance. You will support ML frameworks and compilers across Cloud TPUs/GPUs to train and deploy cutting-edge models. You will collaborate with Google teams and leading researchers to push ML capabilities in a scalable, production-ready environment. You will contribute to design, development, testing, deployment, and maintenance of critical software that accelerates machine learning at scale.
Responsibilities- Develop and optimize AI framework support to run PyTorch on Cloud TPUs and GPUs
- Provide comprehensive ML framework and compiler support on Cloud TPUs/GPUs for training and deployment
- Collaborate with cross-team groups and external researchers to advance PyTorch-in-Cloud capabilities
- Participate in design, development, testing, deployment, and maintenance of software
- Contribute to performance tuning and optimization for large-scale ML workloads
- Bachelor’s degree or equivalent practical experience
- 2 years of software development in one or more programming languages, or 1 year with an advanced degree
- 1 year of experience with ML infrastructure (model deployment, evaluation, data processing, debugging)
- Experience in C++
- collaboration with cross-functional teams
- leadership qualities and initiative
- ability to adapt to fast-paced, research-driven environments
- ML frameworks such as PyTorch/XLA/Pallas/GPU/TPU programming
- compilers
- machine learning optimization
Reference: WJ-747_31295888 · Posted 15 Nov 2026, 08:52 PM