Principal Engineer, SoC NPU Architect

Lab Summary:

The Samsung SoC Architecture Lab explores innovative SoC architecture, key IP blocks (CPU, GPU, NPU), bus / memory subsystem and multimedia subsystems for future Samsung Galaxy products (Smartphones, IoT and future devices) echoing Samsung’s AI for All strategy. We are defining the high-performance SoC architecture development for various Galaxy device lineups. This lab collaborates with Samsung MX headquarter team, key R&D teams around the globe and Samsung's strategic SoC partners to innovate and reinvent technology that will positively impact hundred-millions of people around the world via the Galaxy flagship products.

Position Summary:

We are looking for a NPU Architect for next generation SOCs. This is a highly visible hands-on role leading individual and team contributions to NPU sub-system architecture, interface, performance and power tradeoffs. Responsible for anywhere from early-stage architectural exploration, research work focusing on machine learning, defining new features, performance modeling, Hardware-Software co optimization.

Position Responsibilities:

• Propose innovative NPU Architectural and micro-architectural features to boost PPA (Performance, Power and Area) on various targeted workloads in next generation SOCs
• Perform high-level performance modeling/simulation and analysis of NPU features, applications, benchmarks, and complex uses case
• Develop cycle accurate and analytical NPU performance model and correlate with Silicon
• Deliver architecture/microarchitecture proposals and specifications to the design team and articulate them effectively across audiences ranging from hardware & software engineers to architecture community peers, and to technology leadership
• Analyzing benchmarks, application workloads, and performance simulation results for architecture and microarchitecture optimizations
• Map AI use cases to NPU.
• Optimize and co-design of SW and HW to develop new architectures for low power and high performance
• Collaborate across teams to bring architecture proposals to fruition across the SOC, Driver, OS and System through detailed documentation

Required Skills:

• Bachelor’s, Masters or PhD degree in Electrical, Computer Science, related Science or equivalent combination of education, training and experience
• 16+ years of experience in NPU architecture & development
• Hands on experience in developing cycle accurate performance and functional models
• Good knowledge and hands on experience in in CNN and LLM models
• Proficient programming skills in Python, System C and/or C++
• Experience in ML frameworks (TensorFlow, PyTorch, ONNX)
• Experience in micro architecture of SoC, interface subsystems, logic modules
• Self-motivated problem-solver with an ability to work well in a team

Special Attributes:

• Experience in performance/power modeling using Platform Architect
• Experience in mapping AI use cases to IPs
• Experience with hardware memory managements units, prefetching or memory subsystems
• Experience with multi-core systems and memory coherency
• Experience in ML model optimization on GPU Architecture
• Highly motivated with good verbal and written communication skills