Назад
Company hidden
2 дня назад

Senior NPU Architect (AI Accelerator Architecture)

280 000$
Формат работы
remote (только USA)
Тип работы
fulltime
Грейд
senior
Английский
b2
Страна
US
Вакансия из списка Hirify.GlobalВакансия из Hirify Global, списка международных tech-компаний
Для мэтча и отклика нужен Plus

Мэтч & Сопровод

Для мэтча с этой вакансией нужен Plus

Описание вакансии

Текст:
/
TL;DR
Senior NPU Architect (AI Accelerator Architecture): Designing a new class of AI accelerator based on precise analog in-memory computing, translating AI workload and software requirements into architecture, microarchitecture, and implemented models with an accent on on-chip networks, memory orchestration, and hardware execution efficiency. Focus on implementing architectural features in C++ simulations, analyzing compute utilization and data movement, developing verification strategies, and optimizing execution for LLMs, diffusion models, and CNNs.

Location: Remote within the United States

Salary: Up to $280,000 USD base annually

Company

Work with a semiconductor startup developing commercial AI accelerators based on precise analog in-memory computing after raising a $100M Series B.

What you will do

  • Translate workload and software requirements into architecture, microarchitecture, and implemented models.
  • Define specifications and evaluate architectural tradeoffs for in-memory compute, on-chip networks, and memory orchestration.
  • Implement architectural features in a C++ functional simulation framework.
  • Analyze compute utilization, memory bandwidth, and data movement.
  • Develop verification strategies and guide design decisions for LLM, diffusion, and CNN workloads.
  • Collaborate with hardware, software, and verification engineers.

Requirements

  • PhD with 2–8 years of relevant experience, or a strong MS researcher or engineer with substantial implemented architecture projects.
  • Relevant background in AI accelerator, GPU, or CPU architecture.
  • Ability to explain personally built designs, design choices, and how AI workloads execute on hardware.
  • Experience with C++ and Python, digital-design fundamentals, and HDL familiarity.
  • Prior analog in-memory computing experience is not required.

Culture & Benefits

  • Remote role limited to the United States.
  • Hands-on collaboration across hardware, software, and verification engineering.
  • Opportunity to shape future AI accelerator hardware and its workload execution model.

Будьте осторожны

Если работодатель просит войти в их систему, используя iCloud/Google, прислать код/пароль, запустить код/ПО, не делайте этого. Это мошенники. Обязательно жмите «Пожаловаться» или пишите в поддержку. Подробнее в гайде →