Parallel Software Developer

Crossing Hurdles

Position:

Role Responsibilities:

• Accelerate numeric and simulation kernels using GPU/CPU parallelism.
• Optimize workloads through memory-hierarchy tuning and communication reduction strategies.
• Scale large computational workflows across distributed clusters using MPI, NCCL, and Slurm.
• Profile, benchmark, and tune performance using tools like

  nvprof

  and

  nsys

  .
• Build reproducible HPC pipelines using Python, NumPy, and SciPy.
• Integrate parallel computing optimizations into production AI and scientific systems.
• Collaborate with researchers and engineers to enhance performance across large-scale computational workloads.
  

Requirements:

• Background in computer science, high-performance computing, or applied mathematics.
• Hands-on experience with GPU/CPU parallel programming (CUDA, OpenMP).
• Strong understanding of distributed execution frameworks (MPI, NCCL, Slurm).
• Proficiency in Python with scientific computing libraries (NumPy, SciPy).
• Experience profiling and optimizing workloads using tools like

  nvprof

  and

  nsys

  .
• Strong understanding of memory hierarchy, communication overhead, and scalability constraints.
• Interest in accelerating AI training, simulations, and scientific computations through HPC techniques.
  

Key Domains:

Applicants with strengths in the following domains are highly encouraged to apply:

• High-Performance Computing (HPC):

  Parallel programming, cluster execution, GPU/CPU optimization

• Distributed Systems:

  MPI, NCCL, Slurm, cluster scaling

• Scientific Computing:

  Numeric kernels, simulations, Python-based HPC workflows

• Performance Engineering:

  Profiling, benchmarking, memory tuning

• AI Acceleration:

  HPC methods for AI training and large-scale models

Application Process:

• Apply for the job role
• Await the official message/email from our recruitment team (typically within 1-2 days)