NVIDIA
Senior SoC Subsystem and I/O Architect - LPU
$184K–$356K
US, CA, Remote +2 · Senior · Full-time · Posted Aug 6, 2026
- Location
- US, CA, Remote
US, TX, Remote
US, MA, Remote - Experience
- 8+ years in SoC, subsystem, I/O, interconnect, GPU/CPU, accelerator, or high-performance systems architecture.
- Education
- BS, MS, or PhD in Electrical Engineering, Computer Engineering, Computer Science, or equivalent experience.
About the job
Defines and validates high-level SoC subsystem and I/O architecture for NVIDIA LPU products, from product requirements through specifications, models, and cross-team implementation.
What you'll do
- Translate product requirements into subsystem architecture and detailed IP specifications, covering I/O, memory, firmware, boot, safety, and power.
- Build or guide functional and architectural models to validate subsystem behavior and IP interactions before RTL or silicon is available.
- Drive architecture tradeoffs and coordinate reviews and decisions across IP, firmware, software, RTL, verification, platform, and post-silicon teams.
What they ask for
- Strong understanding of SoC architecture and uncore subsystem development, including I/O and interconnect cores.
- Experience translating product requirements into architecture requirements and IP/subsystem specifications.
- Experience crafting functional, architectural, golden, or C/SystemC models from architecture specifications.
- Knowledge of memory ordering, coherency, address translation, interrupts, virtualization, MMUs/IOMMUs, boot, security, RAS, debug, and power-management architecture.
Nice to have
- Hands-on experience with LPU, GPU, CPU, AI accelerator, chiplet, or multi-die SoC architecture.
- Understanding of deep learning workloads, numerics, memory architecture, and bandwidth/latency tradeoffs.
Pay
184,000 USD - 287,500 USD (base salary); 224,000 USD - 356,500 USD (base salary)
Tools
PCIe/CXL, NVLink/NVLink-C2C, UCIe, AXI/CHI, NoC fabrics, C++, SystemC, Python
AI focus
Architects SoC subsystems for AI and high-performance computing products, including LPU products. NVIDIA also states that it uses AI tools in its recruiting processes.
Model-written summary · the original posting is authoritative