Staff Engineer, AI System Architect (Hardware)

Samsung Semiconductor•San Jose, CA

1d•Onsite

About The Position

The Architecture Research Lab (ARL) focuses on addressing fundamental system-level bottlenecks in modern AI, particularly in memory capacity/bandwidth and system-scale communication. By leveraging Samsung’s world-class memory technologies, ARL explores and defines next-generation AI system architectures that deliver step-function improvements in performance, efficiency, and scalability. We are seeking a Senior Staff AI System Architect who will play a key role in bridging AI workloads, system architecture, and hardware design. In this role, you will develop system-level performance models, drive architecture-level design decisions, and propose forward-looking AI system architectures that shape Samsung’s long-term AI platform strategy.

Requirements

Ph.D. in Computer Science, Electrical Engineering, or a related field, with 5+ years of experience in system architecture for large-scale computing platforms, with a strong focus on AI workloads.
Proven hands-on experience developing analytical and event-driven simulation models for system-level performance evaluation.
Deep understanding of AI system hardware architectures, including compute, memory hierarchies, and high-performance interconnects.
Strong knowledge of modern and emerging AI workloads, including LLMs, DLRMs, and large-scale training and inference systems.
Demonstrated ability to translate workload characteristics and modeling results into actionable architectural design decisions.
Proficiency in Python, C++, and PyTorch for modeling, analysis, and experimentation.
Excellent written, verbal, and presentation communication skills, with the ability to influence technical direction across teams.
A collaborative mindset, intellectual curiosity, and resilience in tackling complex, open-ended system-level challenges.
Inclusive, adapting style to the situation and diverse global norms of people.
Approach challenges with curiosity and resilience, seeking data to help build understanding.
Collaborative, building relationships, humbly offering support and openly welcoming approaches.

Responsibilities

Conduct system-level architectural research for next-generation AI systems, spanning compute, memory, and interconnect/network subsystems.
Develop and maintain analytical and simulation-based system modeling frameworks to evaluate AI workloads and identify performance, scalability, and efficiency bottlenecks at rack- and system-scale.
Analyze representative and emerging AI workloads (e.g., LLMs, DLRMs, and future AI models) to derive architecture requirements and trade-offs across compute, memory, networking, and power.
Drive architecture-level design decisions through quantitative modeling, design-space exploration, and performance/power projections.
Perform comparative studies of alternative system architectures, reporting performance and performance-per-watt metrics to guide strategic technology choices.
Collaborate closely with cross-functional teams in hardware architecture, memory, interconnect, and system engineering to align modeling insights with implementation realities.
Communicate architectural insights and recommendations through clear technical presentations and documentation.