Senior HPC Application Engineer Full-time

Senior HPC Application Engineer

Young and research-intensive, Nanyang Technological University, Singapore (NTU Singapore) is ranked among the world's top universities.

The High-Performance Computing Centre (HPCC) was established in 2010 to support the needs of large-scale and data-intensive computation at the Nanyang Technology University, Singapore (NTU).

For more than 10 years, HPCC has grown and achieved recognition for energy efficiency, pioneering technologies and innovation. Located at the North Spine of NTU, HPCC provides resources, services, and expertise to NTU research and teaching community.

We are seeking a dynamic and enterprising individual to join us as a Senior HPC Application Engineer.

Key Responsibilities

• Port scientific applications to GPU (e.g., using CUDA, HIP, OpenACC) or optimize for multi-core CPUs or optimization of AI applications.
• Optimize memory access patterns, data locality, and compute kernels.
• Profile and tune performance across heterogeneous systems.
• Conduct benchmarking across CPU and GPU nodes using representative AI and scientific workloads.
• Analyze performance metrics and identify bottlenecks.
• Work closely with researchers to understand application needs and provide technical guidance.
• Develop documentation, training materials, and reusable code templates.

Requirements

• Bachelor's degree in Computer Engineering, Sciences or its equivalent.
• Hands-on experience with CUDA, OpenCL, C/C++, and Python.
• Working knowledge for parallel programming models (MPI, OpenMP, CUDA, SYCL)
• Proven track record in GPU driver development, performance profiling, and debugging.
• Familiarity with Linux kernel, Windows WDDM, or embedded systems is a plus.
• Experience working on GPU-accelerated applications in domains like gaming, machine learning, scientific computing, or graphics rendering.
• Exposure to cloud GPU platforms (e.g., AWS, GCP) or energy-efficient computing is increasingly.
• GPU Programming: Expertise in CUDA, OpenCL, C/C++, and Python.
• Architecture Knowledge: Deep understanding of GPU internals, memory hierarchy, and parallel computing.
• Performance Tuning: Experience with profiling tools, debugging, and optimization techniques.
• Domain Tools: Familiarity with graphics APIs (OpenGL) or ML frameworks (TensorFlow, PyTorch)
• Strong problem-solving, communication, and teamwork abilities.

We regret that only shortlisted candidates will be notified.