Senior NPU Architect: AI Accelerator & In-Memory Compute

EnCharge AI; United States; Remote; USD 180,000 - 240,000; Full time

Job summary

EnCharge AI is seeking a Senior NPU Architect to join our Architecture Team. You will collaborate with NPU and system architects, hardware engineers, and software teams to define and optimize features in our AI accelerator based on in-memory computing technology.

You will model architecture in C++ and contribute to performance analyses for workloads like LLMs, diffusion models, CNNs, and more, helping shape future generations of architectures in a fast-growing edge-to-cloud platform.

Qualifications

• BS or MS in EE/CS or related field with 5-8 years of relevant experience.
• Preferred: PhD in a related field with 2-4 years of relevant experience.
• 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; HDL experience is a plus.

Responsibilities

• Define and develop the spec, architecture, and micro-architecture of key architecture modules based on workloads and deployment flow.
• Model key modules in C++ simulation framework to ensure functional implementation of features.
• Collaborate with design verification to test architecture features within modules.
• Analyze performance and efficiency of architecture modules for workloads; guide decisions.
• Stay up to date with AI workloads trends to shape future architectures.

Skills

Computer architecture

Digital design

Micro-architecture

AI/ML algorithms

C/C++

Python

Verilog

SystemVerilog

Education

BS/MS in EE/CS

PhD in related field

Tools

Verilog

SystemVerilog

Job description

EnCharge AI is seeking a Senior NPU Architect to join our Architecture Team. You will collaborate with NPU and system architects, hardware engineers, and software teams to define and optimize features in our AI accelerator based on in-memory computing technology.

You will model architecture in C++ and contribute to performance analyses for workloads like LLMs, diffusion models, CNNs, and more, helping shape future generations of architectures in a fast-growing edge-to-cloud platform.