Come join us to accelerate deep learning networks on Az1/Az2/Az3 Neural Edge processors and beyond. We deliver solutions to offload Speech, Computer Vision and Generative AI workloads on a range of devices from Blink Smart Home camera to Echo Show line of products.
We are looking for a talented and passionate engineering leader to manage an exciting technology creation team at Amazon. You will have an enormous opportunity to lead a team that makes a large impact on the design, architecture, and implementation of deep learning technologies embedded into consumer products used every day, by people you know. The position provides a unique opportunity to guide your team's contributions and make an impact from hardware design stage followed by pre and post silicon development as well as productizing it on consumer devices, while supporting both current and future compiler and ML accelerator architectures.
In this role you will lead a team working alongside partner science teams to develop the compiler infrastructure and lower deep learning workloads to heterogeneous device backends. You will also drive partnerships with peer science teams to innovate on model quantization and compression techniques for efficient execution on hardware, ensuring scalability across current and future generations of ML accelerator architectures.


Key job responsibilities
Lead and manage a team of software engineers developing compiler infrastructure for deep learning accelerators
· Drive technical vision and strategy for the deep learning compiler stack
·. Manage project planning, execution, and delivery of compiler solutions
· Build and maintain strong partnerships with hardware, software, applied science and product teams
·. Develop and mentor team members, manage performance, and drive career growth
· Drive hiring and team building initiatives
· Architect and oversee development of software stack for deep learning accelerator
· Lead design reviews, API development, and documentation efforts
· Ensure successful delivery of deep learning workloads on heterogeneous device backends