Andes Deep Learning Accelerator (AnDLA) is a highly efficient and cost-sensitive AI solution for edge devices and endpoints. AnDLA features hardwired processing units for matrix multiplication, convolution, pooling functions, and more in the future.
Your team designs and builds the AnDLA softcore IP. You lead your team from micro-architecture, design, verification, and documentation. You collaborate with SW engineers to optimize the programming sequence from the scheduling perspective. In this role, you help grow and manage a growing deep learning R&D team.
===================
Responsibilities
===================
AnDLA design lead.
Define the AnDLA architecture as well as the system architecture with AndesCore processors for deep learning applications.
Explore various trade-offs of the AnDLA design in terms of performance, power, energy, and area.
Work with SW engineers to optimize scheduling results.
Work with AI compiler engineers to optimize compilation results.