Google welcomes people with disabilities.
Minimum qualifications:
- Bachelor's degree in Computer Science, related technical field, or equivalent practical experience
- 5 years of experience building ML models or reducing ML model compute and memory costs (e.g., quantization, sparsification, pruning)
- 3 years of experience working with Machine Learning or Artificial Intelligence
Preferred qualifications:
- PhD in Computer Science or related field
- Experience working on ML accelerators or other accelerators such as a GPU or DSP
- Experience with JAX and/or TensorFlow
- Experience with training ML models
- Experience with optimizing ML models
- Understanding of performance implications of processor architecture from a software point of view
About the job
Google's mission is to organize the world's information and make it universally accessible and useful. Our Devices and Services team combines the best of Google AI, Software, and Hardware to create radically helpful experiences for users. We research, design, and develop new technologies and hardware to make our user's interaction with computing faster, seamless, and more powerful. Whether finding new ways to capture and sense the world around us, advancing form factors, or improving interaction methods, the Devices and Services team is making people's lives better through technology.
Responsibilities
- Develop and optimize ML models for EdgeTPU, for example in a Pixel phone.
- Collaborate with ML model optimization teams to develop, enhance, and maintain model exploration tools.
- Collaborate with ML model developers to integrate their authoring and training workflows into the developed tooling.
- Collaborate with teams across Google to enable experiences to effectively utilize the EdgeTPU.