Skip to content
Software Engineer – TPU Performance – Hardware/Software Co-Design
| Company | Google |
|---|
| Location | Sunnyvale, CA, USA |
|---|
| Salary | $141000 – $202000 |
|---|
| Type | Full-Time |
|---|
| Degrees | Bachelor’s |
|---|
| Experience Level | Junior, Mid Level |
|---|
Requirements
- Bachelor’s degree or equivalent practical experience.
- 2 years of experience with software development in one or more programming languages, and with data structures/algorithms.
- 2 years of coding experience in one or more of the following languages: C, C++, Java, or Python.
- 2 years of experience testing, maintaining, or launching software products, and 1 year of experience with software design and architecture.
Responsibilities
- Work within the ML Accelerators Roadmap and Co-Design team and collaborate with TPU compiler team, TPU hardware designers and Google-Deepmind ML researchers, to drive ML system Hardware and Software co-design and optimizations.
- Explore and define future ML accelerator system and chip architecture with objective and data-driven insights.
- Enable the peak performance of future ML systems with full stack ML Hardware and Software (HW-SW) co-design.
- Establish understanding on latest business-critical production ML models (Large-language models, large embedding models etc.) to inform optimizations of model architecture, software system and hardware architecture.
Preferred Qualifications
- Experience with development of architecture simulator and micro-architecture.
- Experience with LLMs and ML frameworks and compilers.
- Experience with ML algorithm and performance optimizations.
- Knowledge of computer architecture, TPUs or other accelerators.
- Excellent communication skills.
