remote
Staff Software Engineer, TPU Performance - Google
Software Engineer
Seeking a Staff Software Engineer to optimize TPU performance for machine learning workloads. This role involves deep technical analysis, coding in C++ and Python, and contributing to cutting-edge hardware acceleration.
About the role
Key Responsibilities
- Analyze and optimize the performance of Tensor Processing Units (TPUs) for machine learning applications.
- Develop and implement performance improvements using C++ and Python.
- Collaborate with hardware and software teams to identify and resolve performance bottlenecks.
- Contribute to the design and architecture of high-performance machine learning systems.
- Profile and benchmark software and hardware interactions to ensure optimal efficiency.
- Mentor junior engineers and share expertise in performance tuning.
Requirements
- Proven experience in software engineering with a focus on performance optimization.
- Strong proficiency in C++ and Python programming languages.
- Experience with machine learning frameworks and hardware accelerators (e.g., TPUs).
- Deep understanding of distributed systems and parallel computing concepts.
- Ability to analyze complex performance issues and develop effective solutions.
Skills
cpythonmachine learning