Principal GPU/NPU AI System Architect

Advanced Micro Devices, Inc.
Austin, TX, United States
13 days ago
Apply on www.careerarc.com
Prepare application

Role details

Contract type
Permanent contract
Employment type
Full-time (> 32 hours)
Experience level
Expert
Working hours
Regular working hours

Tech stack

Artificial Intelligence Systems Engineering Compilers Computer Engineering Microarchitecture Data Centers Firmware Sensor Fusion System Software Pytorch Delivery Pipeline IT Architecture
+7 more
AI Platforms Information Technology Low Latency Deployment Automation ONNX (Open Neural Network Exchange) Format TensorRT Stream Analytics

Job description

WHAT YOU DO AT AMD CHANGES EVERYTHING

At AMD, our mission is to build great products that accelerate next-generation computing experiences-from AI and data centers, to PCs, gaming and embedded systems. Grounded in a culture of innovation and collaboration, we believe real progress comes from bold ideas, human ingenuity and a shared passion to create something extraordinary. When you join AMD, you’ll discover the real differentiator is our culture. We push the limits of innovation to solve the world’s most important challenges-striving for execution excellence, while being direct, humble, collaborative, and inclusive of diverse perspectives. Join us as we shape the future of AI and beyond. Together, we advance your career.

THE ROLE:

The AI Architect will define and drive end-to-end AI system architecture for embedded and edge platforms, with deep expertise in GPU/NPU micro-architecture, AI software stacks, and model behavior. This role bridges silicon capabilities, system software, and AI models, enabling performant, power-efficient, and safe AI deployments across robotics, automotive, and industrial markets. The architect will own technical solutioning from model selection through deployment, working closely with silicon, compiler, software, and product teams, and will represent the AI architecture vision with customers and partners.

THE PERSON:

We are seeking a senior AI systems architect with deep expertise across GPU/NPU architecture, AI software stacks, and model behavior. This individual operates at the intersection of silicon, system software, and applied AI - translating real-world robotics, automotive, and industrial workloads into scalable, production-ready AI platform architectures.

The ideal candidate combines hardware-aware AI model understanding with embedded deployment experience, and can drive full-stack architectural trade-offs across performance, power, memory, safety, and lifecycle constraints. They are technically hands-on when needed, yet comfortable influencing silicon roadmaps, guiding cross-functional teams, and representing architectural strategy with customers and ecosystem partners.

This is a high-impact technical leadership role requiring strong architectural judgment, cross-functional influence without direct authority, and the ability to bridge research, productization, and long-term platform evolution.

KEY RESPONSIBILITIES:

GPU / NPU Architecture & HW-SW Co-Design

  • Develop deep architectural understanding of GPU, NPU, and heterogeneous SoC designs, including memory hierarchies, interconnects, scheduling, and power/performance trade-offs.
  • Guide HW-SW co-optimization strategies for AI workloads across vision, perception, planning, and control.
  • Influence silicon and platform roadmaps using model-driven architectural insights from robotics, automotive, and industrial workloads.
  • Collaborate across silicon, system engineering, software, thermal/mechanical, security, and product teams.
  • Technically lead internal AI engineers and work closely with partners, ISVs, and customers.
  • Act as a technical authority and mentor, influencing architecture decisions without direct reporting authority.

  • Architect AI solutions with strong understanding of model internals (CNNs, Transformers, multi-modal models, sensor fusion, perception stacks).
  • Evaluate and map model characteristics (latency, memory bandwidth, precision, sparsity) onto GPU/NPU execution.
  • Drive model optimization strategies (quantization, pruning, distillation, compilation flows) aligned with embedded constraints.

Model-Aware AI System Architecture

  • Software Stack & Deployment Solutioning

  • Define and optimize AI software stacks spanning:
  • Frameworks (PyTorch, ONNX, TensorRT-like runtimes)
  • Compilers, graph optimizers, and runtime schedulers
  • Drivers, firmware, and OS integration

Lead solutioning for edge and embedded deployment, including OTA updates, lifecycle management, and production-grade robustness.

Ensure scalability from prototype * production * long-term maintenance.

Domain-Focused Architecture Leadership

  • Robotics: perception, localization, SLAM, manipulation, real-time decision pipelines.
  • Automotive: ADAS, autonomous perception, sensor fusion, safety-critical AI execution.
  • Industrial: vision inspection, predictive maintenance, autonomous systems, real-time analytics.
  • Translate domain use-cases into architectural requirements and reusable platform capabilities.

PREFERRED EXPERIENCE:

  • Deep expertise in GPU and/or NPU architecture and execution models.
  • Strong hands-on experience with AI models and inference pipelines, not just framework usage.
  • Proven background in embedded / edge AI systems.
  • Strong understanding of hardware-aware model optimization techniques.
  • Experience in robotics, automotive, or industrial AI domains.
  • Ability to translate customer problems into scalable architectural solutions.
  • Motivating leader with good interpersonal skills; cross-functional & external leadership

ACADEMIC CREDENTIALS:

Bachelor’s or Master’s degree in Computer Science, Computer Engineering, Electrical Engineering, or equivalent

LOCATION: Austin, TX or San Jose, CA

This role is not eligible for visa sponsorship.

LI-BW2

LI-HYBRID

Benefits offered are described: AMD benefits at a glance.

AMD does not accept unsolicited resumes from headhunters, recruitment agencies, or fee-based recruitment services. AMD and its subsidiaries are equal opportunity, inclusive employers and will consider all applicants without regard to age, ancestry, color, marital status, medical condition, mental or physical disability, national origin, race, religion, political and/or third-party affiliation, sex, pregnancy, sexual orientation, gender identity, military or veteran status, or any other characteristic protected by law. We encourage applications from all qualified candidates and will accommodate applicants’ needs under the respective laws throughout all stages of the recruitment and selection process.

AMD may use Artificial Intelligence to help screen, assess or select applicants for this position. AMD’s “Responsible AI Policy” is available here.

This posting is for an existing vacancy.

Requirements

We are seeking a senior AI systems architect with deep expertise across GPU/NPU architecture, AI software stacks, and model behavior. This individual operates at the intersection of silicon, system software, and applied AI - translating real-world robotics, automotive, and industrial workloads into scalable, production-ready AI platform architectures.

The ideal candidate combines hardware-aware AI model understanding with embedded deployment experience, and can drive full-stack architectural trade-offs across performance, power, memory, safety, and lifecycle constraints. They are technically hands-on when needed, yet comfortable influencing silicon roadmaps, guiding cross-functional teams, and representing architectural strategy with customers and ecosystem partners.

This is a high-impact technical leadership role requiring strong architectural judgment, cross-functional influence without direct authority, and the ability to bridge research, productization, and long-term platform evolution., * Deep expertise in GPU and/or NPU architecture and execution models.

  • Strong hands-on experience with AI models and inference pipelines, not just framework usage.
  • Proven background in embedded / edge AI systems.
  • Strong understanding of hardware-aware model optimization techniques.
  • Experience in robotics, automotive, or industrial AI domains.
  • Ability to translate customer problems into scalable architectural solutions.
  • Motivating leader with good interpersonal skills; cross-functional & external leadership

ACADEMIC CREDENTIALS:

Bachelor’s or Master’s degree in Computer Science, Computer Engineering, Electrical Engineering, or equivalent

About the company

At AMD, our mission is to build great products that accelerate next-generation computing experiences-from AI and data centers, to PCs, gaming and embedded systems. Grounded in a culture of innovation and collaboration, we believe real progress comes from bold ideas, human ingenuity and a shared passion to create something extraordinary. When you join AMD, you’ll discover the real differentiator is our culture. We push the limits of innovation to solve the world’s most important challenges-striving for execution excellence, while being direct, humble, collaborative, and inclusive of diverse perspectives. Join us as we shape the future of AI and beyond. Together, we advance your career.

Apply for this position

This job is hosted externally. Click below to view the full posting and apply.

Apply on www.careerarc.com
Prepare application

Good distractions

Talks and stories from around this role — technically off-topic, practically not.

2:19 min

Orchestrating over-the-air firmware updates for vehicle modules

Denis Grahovac · World Congress 2021

4:52 min

Essential phases in building and refining language models

Anshul Jindal Anshul Jindal +1 · World Congress 2025

2:35 min

Preventing remote code execution in PyTorch models

Balázs Kiss · World Congress 2023

2:33 min

Architecting CUDA and the AI software stack

Michael Kagan Michael Kagan +1 · World Congress 2026 Europe

2:20 min

Utilizing custom firmware for variable torque manipulation

Daniel Meilak Daniel Meilak +1 · World Congress 2026 Europe

2:32 min

Core libraries driving inference engines and multi-GPU networking

Adolf Hohl Adolf Hohl · World Congress 2024

Videos

See all

Related articles

See all