Frenxt

Description: About the Company A global leader in electronic design automation software, this organization empowers innovators in the microchip and semiconductor industries to push the boundaries of technology. With cutting-edge solutions that enable optimal circuit design and semiconductor development, the company serves as a trusted partner to organizations driving the future of electronics and computational hardware. About the role We are seeking a GPU Architecture and Performance Expert to assess and advise on GPU porting strategy for our product, specifically evaluating whether GPU acceleration can deliver material performance improvements for our integer-based algorithms. Responsibilities Conduct comprehensive technical assessment of the product's native algorithms to determine viability and efficiency gains from GPU acceleration Analyze integer-based computational workloads and specialized solvers to evaluate whether GPU architectures can provide meaningful performance improvements Provide strategic recommendations on GPU investment decisions, including cost-benefit analysis and expected acceleration potential for specific algorithm types Deliver data-driven insights on which computations are strong candidates for GPU porting versus those that should remain CPU-optimized Guide technical teams on implementation strategies if GPU acceleration proves viable, including architecture selection and optimization approaches Required Skills (All Must Haves) Deep expertise in GPU architectures (CUDA, OpenCL, or similar) with proven experience evaluating workload suitability for GPU acceleration Strong understanding of the performance characteristics of integer-based algorithms on GPU versus CPU architectures Demonstrated experience analyzing and optimizing computational workloads for heterogeneous computing environments Proficiency in performance profiling, benchmarking, and cost-benefit analysis for hardware acceleration decisions Solid foundation in parallel computing principles and understanding of when GPU acceleration provides material benefits Open Positions: 1 Detailed Required Skills: Algorithms, Architecture, GPU Computing, NVIDIA CUDA Job Timezone: America/New_York Work Type: Remote Estimated Length: 4-8 weeks