NPU Architect

EnCharge AI

8d

About The Position

EnCharge AI is seeking an NPU Architect to join our Architecture Team. In this role, you will collaborate with a team of NPU and system architects, hardware engineers, and software teams to define and optimize key features in our AI accelerator with the cutting-edge in-memory computing technology. You will contribute to designing a holistic AI HW/SW solution that has the best performance and efficiency on the market for the latest AI/ML workloads, such as LLMs, diffusion models, CNNs, and more.

Requirements

• BS/MS/PhD in EE, CS, or a related field
• Understanding in computer architecture, digital design, and micro-architecture concepts
• Familiarity with AI/ML algorithms, frameworks, and workloads
• Programming experience in C/C++ and Python
• Experience with HDLs such as Verilog or System Verilog

Responsibilities

• Define and develop the spec, architecture, and micro-architecture of key architecture modules (such as the in-memory compute unit, on-chip network, and memory orchestration units) based on the requirements of the workloads and software deployment flow
• Contribute to the modeling of aforementioned key architecture modules in our C++ simulation framework to ensure a functional implementation of the features
• Collaborate with the design verification team to deliver a strategy and infrastructure for the testing of the architecture features within said modules
• Work with the software team and other architects to analyze the performance and efficiency of the architecture modules for key workloads, identify performance bottlenecks, and guide architectural decisions
• Stay up to date with the latest trends and research in AI workloads, architectures, and applications to help define a path for the future generations of architectures

What This Job Offers

• Job Type: Full-time
• Education Level: Ph.D. or professional degree