GPU Architect

Job Overview

The GPU Architect will lead the design and optimization of GPU-based high-performance computing architectures tailored for OPC (Optical Proximity Correction) software solutions. They will define HPC solutions that effectively utilize current hardware but prepare us for cutting-edge GPU architectures on the horizon that will impact our industry. As part of their role, they will work closely with HW vendors and customers, demonstrating both technical proficiency and strong communication skills. With product management, they will set a technical roadmap to ensure a solid future of our software on GPU. 

Responsibilities and Duties

• Design and develop methodologies of HPC with in-depth performance characterization and analysis on large multi-GPU and multi-node clusters.  Write software and utility to study the interaction of CPU/GPU software components with all HW (GPU, CPU, Networking) considered in the stack to craft specs and to prototype tooling for measure, visualize.
• Design and develop high quality software (parallel algorithm) of CUDA/ROCm/Hip code to accelerate algorithms in the areas of computation geometry/graphics/signal-image processing.  
• Host, design, and plan collaborative activities with a very dynamic team across multiple time zones in the areas of HPC/network-centric software development team, Database/ML/AI teams, and computational lithography-centric algorithm teams.

Qualifications

• MS/Ph.D. in CS/CE/EE or computation-oriented engineering majors or equivalent
• Substantial experience (10+ years) in EDA HPC field, prior role of tech lead on large-scale software development.  Demonstratable personal establishments in GPU coding/debugging and parallel algorithm development for numerical analysis, computational geometry, graphics, and signal-image processing.
• Actual application/development level experience in computation lithography is a plus.
• Necessary skills include hands-on operation and algorithm-development-oriented skills in Linux-network/cloud computing/CUDA/ROCm/C/C++/Python/ML/AI
• Experience in image and signal processing required that is demonstrated by relevant work or project