13 часов назад
AI/ML ASIC Architect (AI/ML)
194 425 - 322 092$
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Описание вакансии
Текст:
TL;DR
AI/ML ASIC Architect (AI/ML): Developing AI storage solutions and accelerator ASIC architectures integrating GPU/TPU/xPU systems through UCIe, PCIe, and CXL, with an accent on high-bandwidth memory, DMA, and modular I/O subsystems. Focus on architecting memory-efficient LLM training and inference systems, analyzing PPA tradeoffs, and optimizing hardware and software performance across pre-silicon and post-silicon stages.
Location: Milpitas, California, United States; onsite at the Milpitas office
Salary: $194,425–$322,092 per year
Company
develops flash memory, storage solutions, and advanced data-centric hardware products for businesses and consumers.
What you will do
- Drive AI/ML ASIC architecture integrating AI storage with GPU, TPU, NPU, and other xPU accelerators.
- Define PCIe, UCIe, CXL, DMA, NoC, memory-controller, embedded-processor, and FPGA-fabric interactions.
- Create modular I/O subsystem architectures for chiplet, monolithic, and 3D package implementations.
- Author architecture specifications and guide pre-silicon design, verification, emulation, firmware, and post-silicon validation.
- Analyze performance, power, area, endurance, and schedule tradeoffs with customers and cross-functional teams.
- Characterize LLM workloads and optimize memory-efficient training and inference systems, including KV cache, Flash Attention, MoE, pruning, quantization, batching, and speculative decoding.
Requirements
- Bachelor’s, master’s, or PhD in Computer Engineering or Electrical Engineering.
- 15+ years of hands-on architecture experience authoring specifications.
- Strong ASIC, SoC, or I/O subsystem architecture experience with PCIe, UCIe, CXL, and DMA engines.
- Experience with GPU/TPU/xPU accelerator systems, high-bandwidth memory hierarchies, large-scale ML systems, and LLM training or inference.
- Knowledge of computer and graphics architecture, microarchitecture, performance engineering, ARM processors, and AXI interconnects.
- Experience coordinating with hardware, software, firmware, design, verification, simulation, emulation, and validation teams.
Nice to have
- Experience with NVLink, UAL, InfiniBand, RDMA, CUDA, NVMe, NAND flash, or distributed training systems.
- Expertise in transformer architectures, attention mechanisms, model parallelism, and GPU memory optimization.
Culture & Benefits
- Full-time employee position with exempt status.
- Medical, dental, vision, life, accident, disability, flexible spending, and health savings benefits.
- Paid vacation and sick leave, tuition reimbursement, transit benefits, and an employee assistance program.
- Eligibility for short-term incentives, long-term incentives, restricted stock units, employee stock purchase plans, and a 401(k) savings plan may apply.
- Inclusive workplace with equal-opportunity and accessibility commitments.
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 →
Похожие вакансии
12 часов назад
I/O Architect (ASIC)
15 часов назад
Principal Network Architect (AI Hardware)
250 000 - 300 000$
11 часов назад
AI Architect (Platform Architect)
280 500 - 313 500$
3 часа назад
Principal Forward Deployed Engineer (AI)
175 000 - 240 000$
18 часов назад
Silicon Domain Architect (FPGA Security)
118 900 - 178 000$
18 часов назад
Datacenter System Software Architect (RISC-V/Linux)
231 444 - 282 876$