Job Title : HPC Engineer Performance Engineering & Benchmarking
Experience : 3+ Years
Location : Bangalore
Employment Type : Full-time
Joining : Immediate
About the Role :
We are looking for a highly skilled HPC Engineer to join our Performance Engineering and Benchmarking team. The role involves performance analysis, optimization, and benchmarking of large-scale high-performance computing (HPC) applications and systems. The ideal candidate will work closely with developers, researchers, and system architects to improve performance and efficiency across CPU, GPU, and hybrid computing environments.
Key Responsibilities :
- Perform performance analysis, tuning, and optimization of HPC applications using MPI, OpenMP, and CUDA.
- Execute and evaluate performance benchmarks such as HPL, STREAM, OSU, SPEC MPI, and others.
- Utilize profiling and performance analysis tools like Intel VTune, HPCToolkit, ARM MAP, and Perf to identify bottlenecks and enhance performance.
- Conduct comparative studies and benchmarking across CPU, GPU, and accelerator architectures, analyzing interconnect performance and scalability.
- Prepare detailed performance evaluation reports, presenting key findings, insights, and optimization recommendations.
- Work with job schedulers (SLURM, PBS, etc.) to manage and execute HPC workloads efficiently.
- Collaborate with developers and research teams to optimize code for parallel scalability and computational efficiency.
- Support the development and maintenance of automated benchmarking and reporting frameworks.
Required Skills and Experience :
Minimum of 3 years of hands-on experience in HPC performance tuning and benchmarking (no upper limit).Strong programming experience in C or C++ (preferred) or Python for automation and analysis.Practical knowledge of parallel programming models such as MPI and OpenMP.Experience with GPU programming using CUDA or OpenCL.Familiarity with job schedulers like SLURM or PBS.Good understanding of HPC system architecture, including processors, memory hierarchy, interconnects, and accelerators.Ability to work with profiling and performance tools to identify and fix application performance bottlenecks.Good to Have Skills :
Experience with hybrid computing architectures and heterogeneous clusters.Knowledge of containerization tools (Singularity, Docker) in HPC environments.Familiarity with cloud-based HPC platforms (Azure, AWS ParallelCluster, Google Cloud HPC).Exposure to scientific computing workloads or AI / ML performance tuning.Experience in writing technical documentation and performance reports.Additional Information :
Candidates with strong analytical and problem-solving skills will be preferred.This is a hands-on role that requires deep understanding of HPC performance metrics and tuning methodologies.Opportunity to work on cutting-edge HPC systems and collaborate with global teams in optimizing large-scale scientific and engineering workloads.(ref : hirist.tech)