Назад
Company hidden
5 дней назад

AI Accelerator Compute Architect (AI)

120 000 - 275 000$
Формат работы
hybrid
Тип работы
fulltime
Грейд
junior/middle/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
AI Accelerator Compute Architect (AI): Defining compute architecture for next-generation GenAI accelerators, including instruction sets, connectivity, memory, and control, with an accent on workload-driven architecture and silicon performance. Focus on analyzing latency, throughput, utilization, scalability, and power-area-performance tradeoffs while supporting first-silicon bring-up and post-silicon debug.

Location: Hybrid role based in the Mountain View, California office, with in-office work Tuesday through Thursday. Exceptional remote arrangements may be considered.

Salary: $120,000–$275,000 for early career, $175,000–$450,000 for mid career, or $275,000–$600,000 for senior career, plus equity.

Company

hirify.global develops vertically integrated hardware and software solutions, from silicon to systems, for training and running large AI models.

What you will do

  • Define the compute architecture for next-generation GenAI accelerators, including ISA, connectivity, memories, and control.
  • Derive architectural requirements from AI use cases and translate workloads into hardware architecture.
  • Conduct analytical and empirical performance analysis.
  • Collaborate with Research, Software, and Hardware teams to balance workloads, programmability, and PPA.
  • Evaluate power, area, and performance tradeoffs with stakeholders.
  • Support test plan reviews, first-silicon bring-up, and post-silicon debug.

Requirements

  • Strong computer architecture background, especially in AI accelerators, GPUs, TPUs, vector processors, SIMD, VLIW, or other parallel compute architectures.
  • Solid microarchitecture fundamentals and first-principles understanding of latency, throughput, utilization, and scalability.
  • Experience translating workloads into hardware architecture.
  • Familiarity with numeric formats, quantization, rounding, and precision-performance tradeoffs.
  • Programming, code optimization, or kernel-level performance analysis experience.
  • Excellent written and verbal communication skills.

Nice to have

  • Experience optimizing performance in large-scale transformer models.

Culture & Benefits

  • Four weeks of accrued PTO, 12 company holidays, and up to three weeks of remote work.
  • Company-subsidized medical, dental, vision, life, HSA, and FSA benefits.
  • Retirement plans with up to 5% company contribution to a 401(k).
  • $1,500 annual professional development budget.
  • Onsite weekday meals and fully covered daily commuting options.
  • Paid parental leave, mental health benefits, and remote home-office setup support.

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 →