remote
Staff Software Engineer, ML Compilers TPU - Google
Software Engineer
Lead the design, implementation, and optimization of machine‑learning compilers for TPU hardware, driving performance improvements across large‑scale systems using C++ and Python.
About the role
Key Responsibilities
- Architect, develop, and maintain high‑performance ML compiler components targeting TPU architectures.
- Design and implement compiler optimizations that improve runtime efficiency and resource utilization.
- Conduct large‑scale performance analysis, profiling, and debugging to identify bottlenecks and guide enhancements.
- Collaborate with cross‑functional teams to integrate compiler features into end‑to‑end ML pipelines and product releases.
- Mentor junior engineers and establish best practices for code quality, testing, and documentation.
Requirements
- 8+ years of professional experience programming in C++ and Python.
- 5+ years of experience shipping complex software products, including testing and launch processes.
- 3+ years of experience in software design and architecture for large‑scale systems.
- Proven expertise in performance analysis, visualization tools, and debugging of high‑throughput workloads.
- Hands‑on experience with state‑of‑the‑art ML compilers and compiler optimization techniques.