HPC Software EngineerSynopsys 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 have spent your career making code run faster on hardware that keeps changing. You understand that the difference between a solver that takes three hours and one that takes twenty minutes is not just compute power, it is how the algorithm maps to memory, how threads communicate, and whether someone thought hard about cache locality six months ago.
You know your way around a profiler and you can read the output without needing to Google every metric.Working across CPU and GPU architectures does not intimidate you. You have debugged MPI communication bottlenecks, wrestled with build systems that span multiple platforms, and written code that has to perform consistently whether it is running on a local workstation or a cloud-based HPC cluster. You think in terms of scalability, not just correctness.Collaboration comes naturally. You can sit with a domain expert who speaks in finite elements and boundary conditions, extract what matters for the code, and turn that into a design spec that your team can actually build from.
You care about the engineers downstream who depend on your work holding up under real-world load, not just passing CI.At Synopsys, you will work on simulation software that shapes how products get designed across industries. The problems are real, the scale is significant, and what you optimize will matter.Design, implement, and optimize parallel programming methods within Ansys Mechanical solver products using MPI, GPU programming models like CUDA, HIP, SYCL, OpenMP, and other HPC frameworksProfile solver performance across CPU and GPU architectures using tools like Intel VTune, NVIDIA Nsight, or similar, and translate findings into actionable performance improvementsBuild and maintain code benchmarking suites that track solver performance across releases and catch regressions before they shipDrive adoption of modular, hardware-agnostic HPC programming models across multiple solver codebases, working with development teams to ensure consistency and reusabilityCollaborate with numerical methods experts to translate complex algorithmic requirements into performant, maintainable software designsSupport procurement, configuration, and management of HPC development and testing platforms, including on-premise clusters and cloud-based environmentsOwn packaging, build system work, and DevOps tooling using CMake, Azure DevOps, Conan, Docker, or CI/CD pipelines to streamline deployment and testing workflowsReduce solve times for engineering simulations used by leading companies across automotive, aerospace, energy, and electronics industriesEnable customers to run larger, more complex models by making solvers scale efficiently across hundreds or thousands of coresAccelerate the adoption of GPU computing in production simulation workflows, unlocking new performance tiers for users with modern hardwareImprove developer productivity across multiple solver teams by building reusable HPC frameworks and shared toolingEnsure performance consistency and reliability across solver releases through rigorous benchmarking and regression testingHelp shape the technical direction of Synopsys simulation products as HPC architectures and customer workloads continue to evolveSupport faster iteration cycles for product development teams by streamlining build, test, and deployment infrastructureMinimum Requirements: Bachelor's degree in Mechanical Engineering, Computational Science, Applied Mathematics, Physics, or related field with 2+ years of experience, or Master' s degree in a related field. PhD preferred.Strong hands-on experience with HPC software design, testing, and deployment in production or research environmentsSolid understanding of data structures, algorithms, and performance considerations in parallel computing contextsProficiency with Git and collaborative development workflows across distributed teamsProficiency in Fortran and C/C++ for performance-critical code developmentExperience with MPI and distributed memory programming modelsExperience with GPU hardware and at least one GPU programming model such as CUDA, HIP, SYCL/ oneAPI, OpenMP, OpenACC, or Kokkos is a strong plusYou can look at a profiler trace and identify the bottleneck without needing three meetings to discuss itYou write code that other engineers can pick up six months later without needing you to explain every design choiceYou ask clarifying questions when a requirement is vague rather than guessing and building the wrong thingYou are comfortable managing your own time across multiple priorities and know when to escalate blockers versus solve them yourselfYou can explain a technical tradeoff between two HPC approaches to a domain expert in terms they care about, not just what the benchmark saysYou stay current with HPC architecture trends and programming models without needing to be told, because you care about building software that will still perform well two hardware generations from nowYou will work within the Ansys Mechanical R&D organization, collaborating with numerical methods experts, solver developers, and other HPC engineers across a geographically distributed team. Your work will directly support the performance and scalability of industry-leading simulation products used by engineers worldwide to design and validate critical systems.We offer a comprehensive range of health, wellness, and financial benefits to cater to your needs.
Our total rewards include both monetary and non-monetary offerings. Your recruiter will provide more details about the salary range and benefits during the hiring process.