Photonic Engineer, Machine Learning Systems, Platforms

Google•Sunnyvale, CA

About The Position

Google’s machine learning systems use the latest and greatest optical transceiver and optical switching technologies to interconnect TPU and GPU processors. As a Photonic Engineer for Machine Learning Systems, you will be responsible for creating next generation optical interconnect technologies for TPU systems. You will work with both the optics team and machine learning team to lay out the optical technology road maps to expand future generation machine learning systems. Our Platforms Infrastructure Engineering team designs and builds the hardware and software technologies that power all of Google's services. Our computational challenges are complex and unique, enabled by custom hardware designed and made in-house. As a hardware engineer, you will design and build the systems that are the heart of the world's largest and most powerful computing infrastructure. You will see those systems from concepts all the way through to high-volume manufacturing. Your work has the potential to shape the machinery that goes into our data centers, affecting millions of Google users.

Requirements

PhD degree in Electrical Engineering, Computer Engineering, Physics, a related field, or equivalent practical experience.
2 years of experience working in a high-speed datacom technical environment.

Nice To Haves

Experience with script writing and test and measurement system automation.
Knowledge of large volume high-speed optical transceiver designs and manufacturing.
Knowledge of data center network and machine learning system architectures.
Proficiency in digital signal processing, high-speed optoelectronic circuit simulation or system modeling.

Responsibilities

Design low-cost and high-speed optical transceivers for data center applications.
Develop optical systems and architectures to expand TPU and GPU superpods.
Identify, research, evaluate emerging photonic networking technologies.
Explore optical technologies to disaggregate and improve the efficiencies of future computing systems.