R&D Engineering, Staff Engineer - 18291

We Are

Synopsys is the leader in engineering solutions from silicon to systems, enabling customers to rapidly innovate AI-powered products. We deliver industry-leading silicon design, IP, simulation and analysis solutions, and design services. We partner closely with our customers across a wide range of industries to maximize their R&D capability and productivity, powering innovation today that ignites the ingenuity of tomorrow.

You Are

You have spent years building computational mechanics software that engineers actually depend on to simulate real-world physics, not toy problems, but production simulations running on massive clusters where performance and accuracy both matter. The difference between code that scales to thousands of cores and code that stalls out is something you understand at the implementation level, not just in theory. You have debugged parallel code that behaves differently on different architectures, and you know how to trace that back to a memory access pattern or a communication bottleneck that no profiler flagged.

Moving work from CPU to GPU is not a buzzword exercise for you. You have done it. You know what translates cleanly and what requires rethinking the algorithm entirely. Designing and extending mechanics software within large Fortran and C codebases is familiar territory. You are comfortable in code that predates modern tooling, and understand the underlying physics and numerical algorithms needed for complex engineering simulations.

OpenMP, MPI, CUDA, OpenACC: these are not resume keywords, they are tools you reach for depending on the problem.

At Synopsys, you will work on LS-DYNA, software that powers crash simulations, structural analysis, and complex dynamics problems across automotive, aerospace, and civil engineering. The team is small, the problems are hard, and what you build will run on some of the largest supercomputers in the world.

What You ll Be Doing

• Develop, optimize, and maintain LS-DYNA solvers and features including CPG, CPM, SPH, DEM, PBM, and EGMM methods
• Contribute to the development of meshfree and finite element methods that balance accuracy, stability, and computational efficiency
• Port SPH and other performance-critical features from CPU to GPU using OpenACC, OpenMP Offload, and CUDA
• Debug and resolve consistency issues in LS-DYNA HYBRID on large-scale customer models where identical inputs must produce identical results regardless of thread count or CPU allocation
• Implement and optimize parallel algorithms using OpenMP and MPI for distributed memory and shared memory systems
• Profile, benchmark, and tune code performance on HPC clusters and GPU-accelerated systems
• Collaborate with ACE engineers and customers to translate requirements into software solutions, diagnosing simulation issues and delivering fixes that hold up in production environments

• Reduce simulation runtime for automotive crash models by porting SPH solvers to GPU, enabling faster design iteration cycles for customers
• Implement novel numerical techniques to enable customers to solve increasingly complex engineering challenges
• Resolve consistency and accuracy issues in HYBRID solvers that directly affect customer confidence in simulation results, ensuring production runs scale reliably across varying compute resources
• Improve scalability of LS-DYNA on next-generation HPC architectures, ensuring the software remains competitive as hardware evolves
• Deliver performance improvements that allow engineers to run larger, more detailed models without requiring additional compute resources
• Contribute to the long-term maintainability and robustness of a codebase used by thousands of engineers worldwide
• Help customers move critical workloads to modern HPC platforms, reducing their time to insight and increasing their R&D throughput

• Bachelor s or Master s in Mechanical Engineering, Civil Engineering, Computer Science, or related field
• 4-5+ years of hands-on experience developing computational mechanics software or HPC simulation tools
• Strong proficiency in Fortran, C, and C++ for performance-critical scientific computing
• Demonstrated experience with parallel…