R&D Engineering, Staff Engineer (Fusion Compiler GPU Acceleration)

About the role

The Fusion Compiler GPU Acceleration team at Synopsys is dedicated to developing industry-first GPU-accelerated digital implementation solutions. This role involves designing, developing, and owning GPU acceleration for engines across the Fusion Compiler R2G flow, including placement, global routing, detail routing, clock tree synthesis, optimization, timing analysis, extraction, legalization, and synthesis.

Responsibilities

• Design, develop, and own GPU acceleration for engines across the Fusion Compiler R2G flow, including placement, global routing, detail routing, clock tree synthesis, optimization, timing analysis, extraction, legalization, and synthesis
• Reformulate complex EDA algorithms to take full advantage of GPU architectures, balancing performance, memory constraints, and numerical accuracy
• Own projects end to end, from requirements gathering and design specification through development, testing, deployment, and direct customer interaction
• Collaborate closely with cross-functional teams including product management, product engineering, and field teams to align acceleration strategies with real customer workflows
• Debug and optimize performance-critical C/C++ code and CUDA kernels across large, multi-component codebases
• Contribute to the ongoing Nvidia and Synopsys GPU acceleration collaboration, helping define what industry-first GPU-accelerated digital implementation looks like

Requirements

• Bachelor's, Master's, or PhD in Computer Science, Electrical Engineering, or a related field with a strong academic record
• For experienced candidates, 3 to 6 years of hands-on experience developing software projects, preferably in EDA, semiconductor, or high-performance computing domains
• For fresh PhD or MS graduates, demonstrated proficiency in C/C++ through coursework, research projects, or publications, with strong foundations in algorithms, data structures, and system design
• Expert or emerging proficiency in C/C++ development, with a track record or academic evidence of delivering robust, scalable solutions
• Experience with CUDA, GPU acceleration, or GPU architecture knowledge is a strong plus but not required
• Research or project experience in areas such as GPU computing, parallel algorithms, computer architecture, or systems programming is highly valued for recent graduates

Qualifications

• You can take a complex EDA algorithm, break it down into parallelizable components, and explain the tradeoffs to a senior architect in two sentences without losing the nuance
• You are comfortable owning a project from concept to customer deployment, navigating ambiguity, shifting requirements, and cross-functional dependencies along the way
• You approach new languages and technologies with curiosity and adaptability, whether that means learning CUDA for the first time or diving into a new corner of the Fusion Compiler codebase
• You are a strong problem solver with a strategic mindset and attention to detail, someone who thinks about edge cases, memory bottlenecks, and long-term maintainability before the first line of code is written
• You are collaborative and eager to learn from others, whether that means pairing with a senior engineer on a tricky kernel optimization or presenting your work to a product management team
• You are resilient in the face of evolving challenges and requirements, and you thrive in environments that demand deep technical rigor and continuous learning

Skills

• Strong proficiency in C/C++ and CUDA
• Experience with GPU architectures and acceleration techniques
• Knowledge of EDA tools and workflows
• Ability to work effectively in a cross-functional team environment
• Excellent problem-solving and debugging skills
• Strong communication and presentation skills

Benefits

• Comprehensive health, wellness, and financial benefits
• Time Away programs
• Family support programs
• Retirement plans
• Competitive salaries