AI Insights
NVIDIA

SoC Product Architect Low Power SoC

NVIDIA · Santa Clara, California, US
full-timestaff (12-20 yrs)Posted 30d ago
Hardware / Silicon EngineeringIC5IC + ManagementOn-site
StackSoC ArchitectureCPU ArchitectureGPU ArchitectureLPDDR5LPDDR6SRAMDVFSPower DomainsPower ManagementLow-Power IOTensorRTTFLiteONNXDL InferencePPA AnalysisMemory Subsystem DesignQoSDRAM Bandwidth AnalysisPRD Development

Summary

Senior SoC/Product Architect role at NVIDIA defining next-generation low-power SoC architecture, owning concept-to-product decisions spanning CPU, GPU, memory subsystems, and power management, while serving as the technical interface between internal teams and customers.

About the role

NVIDIA has been transforming computer graphics, PC gaming, and accelerated computing for more than 25 years. It’s a unique legacy of innovation that’s fueled by great technology—and amazing people.

Today, we’re tapping into the unlimited potential of AI to define the next era of computing. An era in which our GPU acts as the brains of computers, robots, and self-driving cars that can understand the world. Doing what’s never been done before takes vision, innovation, and the world’s best talent. As an NVIDIAN, you’ll be immersed in a diverse, supportive environment where everyone is inspired to do their best work. Come join the team and see how you can make a lasting impact on the world.

We are looking for a senior SoC/Product Architect to define and drive the architecture of next-generation low-power SoCs at NVIDIA. You will own concept-to-product architecture decisions — balancing power, performance, and area — while partnering across CPU, GPU, memory, and IO teams to resolve system-level trade-offs and acting as the technical interface for product requirements and customer feedback.

What you'll be doing: 

  • Define SoC architecture for low-power platforms — CPU, GPU, memory subsystem, low-power IOs, and power management

  • Drive concept-level PPA trade-off analysis and lead the power/performance budget across the SoC

  • Architect the memory subsystem (LPDDR5/6, on-chip SRAM, bandwidth, QoS) for constrained power envelopes

  • Work with DL teams for evaluating and crafting workloads for low power — mapping tasks to constrained hardware; define optimization opportunities in partnership with Nvidia’s DL team.

  • Partner with CPU, GPU, and IO architecture teams inside NVIDIA to align on shared IPs and integration strategies

  • Translate customer requirements into architectural constraints; drive PRD development and requirements triage

  • Engage directly with customers pre-silicon to collect feedback and follow through on product decisions

What we need to see:

  • 12+ years in SoC or product architecture, with deep hands-on knowledge of CPU, GPU, memory, and low-power IO subsystems

  • Master's Degree in Computer Engineering or Electrical Engineering (or equivalent experience)

  • Strong understanding of SoC architecture for low power mobile devices, including DVFS, power domains, and retention strategies

  • Experience with DL inference at the edge (TensorRT, TFLite, ONNX) and mapping workloads to constrained hardware

  • Track record of driving architecture decisions across multi-functional teams in a large organization

Join the future of chip building, you will contribute to innovative products with a dedicated team!

Your base salary will be determined based on your location, experience, and the pay of employees in similar positions. The base salary range is 224,000 USD - 356,500 USD for Level 5, and 272,000 USD - 431,250 USD for Level 6.

You will also be eligible for equity and benefits.

Applications for this job will be accepted at least until March 23, 2026.

This posting is for an existing vacancy. 

NVIDIA uses AI tools in its recruiting processes.

NVIDIA is committed to fostering a diverse work environment and proud to be an equal opportunity employer. As we highly value diversity in our current and future employees, we do not discriminate (including in our hiring and promotion practices) on the basis of race, religion, color, national origin, gender, gender expression, sexual orientation, age, marital status, veteran status, disability status or any other characteristic protected by law.

What you'll do

1Define SoC architecture for low-power platforms covering CPU, GPU, memory subsystem, low-power IOs, and power management
2Drive concept-level PPA (Power, Performance, Area) trade-off analysis and lead power/performance budgeting across the SoC
3Architect memory subsystems (LPDDR5/6, on-chip SRAM) optimizing for bandwidth and QoS within constrained power envelopes
4Partner with DL teams to evaluate and craft workloads for low-power mapping and define optimization opportunities
5Collaborate with CPU, GPU, and IO architecture teams to align on shared IPs and integration strategies
6Translate customer requirements into architectural constraints and drive PRD development and requirements triage
7Engage directly with customers pre-silicon to gather feedback and drive product decisions

Requirements

12+ years of hands-on SoC or product architecture experience spanning CPU, GPU, memory, and low-power IO subsystems
Deep expertise in low-power mobile SoC design including DVFS, power domains, and retention strategies
Proven experience with edge DL inference and workload mapping using TensorRT, TFLite, or ONNX on constrained hardware
Track record of driving architecture decisions and requirements across multi-functional teams in a large organization
Master's degree in Computer Engineering or Electrical Engineering or equivalent industry experience

Nice to have

LPDDR6
Power/performance budget analysis
PRD development
Customer-facing pre-silicon engagement
Workload optimization for constrained hardware

Role overview

Role family
Hardware / Silicon Engineering
Level
IC5 — other
Experience
12–20 years
Type
Hybrid (IC + Management)
Remote policy
On-site
Visa sponsorship
Not offered

Tech stack analysis

FRAMEWORKS
TensorRTTFLiteONNX
TOOLS
LPDDR5LPDDR6SRAMDVFS

Green flags

5 items
Salary range is explicitly disclosed for two levels ($224K–$431.25K), a rare and highly transparent practice in semiconductor hiringcompensation

Discover all 5 green flags for this role

Sign up free →

Benefits breakdown

HEALTH & WELLNESS
Medical benefits (inferred as standard NVIDIA package)
Dental benefits (inferred)
Vision benefits (inferred)

See all benefits organized by category — health, financial, time off & more

Sign up free →

Hiring insights

JD quality
8/10
Urgency
medium
Autonomy
high
Team size
large (15+)

See JD quality score, hiring urgency & team details

Sign up free →

Red flags

PRO4 items
12+ years of experience required — very senior bar may screen out strong candidates with 10–11 years of equivalent depthrequirements

See all 4 red flags — what the JD isn't telling you

Sign up free →

Interview insights

PRO
Rounds
6
Duration
6 wks
Difficulty
very hard
Take-home
No

Get full interview breakdown — rounds, likely topics & prep tips

Sign up free →

Career path

PRO
Next roles
Principal SoC ArchitectDistinguished EngineerVP of Architecture

See where this role leads — full career progression

Sign up free →
About the company

NVIDIA is the world's leading designer of GPUs and AI computing platforms. Its chips power everything from gaming and data centers to autonomous vehicles and scientific research. With a market cap exceeding $2 trillion, NVIDIA's CUDA platform and AI accelerators have become the backbone of the global AI revolution.

HQSanta Clara, CA, USA
Interview difficultyvery hard
Build vs Maintainbuild
Cross-functionalYes