Nvidia Senior AI Platform Engineer 

United States, Texas 

576831018

What you’ll be doing:

• Develop and refine infrastructure for scalable AI workloads optimized specifically for NVIDIA hardware and software.

• Debug and troubleshoot AI jobs running on NVIDIA hardware, leveraging tools such as Nsight, CUDA, and NCCL, to enhance efficiency and performance.

• Collaborate with internal teams, including research, engineering, and data science, to identify areas for efficiency improvement across NVIDIA's AI workloads.

• Optimize AI frameworks, particularly PyTorch, to fully utilize NVIDIA hardware capabilities.

• Create and maintain NVIDIA-specific performance analysis tools and scripts to monitor, assess, and improve AI infrastructure reliability and performance.

• Utilize NVIDIA's profiling tools (e.g., Nsight Systems, Nsight Compute, DCGM) to gain deep insights into bottlenecks and optimize AI task performance on NVIDIA hardware.

What we need to see:

• 8+ years of experience in debugging and optimizing AI workloads on NVIDIA GPUs.

• Bachelor's or Master's degree in Computer Science, Software Engineering or equivalent experience.

• Strong programming experience with Python, especially using AI frameworks like PyTorch, with a focus on NVIDIA optimization.

• Proficiency in C++, with experience in GPU programming, CUDA, and familiarity with NVIDIA-specific optimizations.

• Knowledge of NVIDIA hardware architectures, including GPU memory hierarchy, CUDA cores, and performance tuning techniques.

Ways to stand out from the crowd:

• Experience with large-scale distributed AI systems on cloud or on-premises infrastructure using NVIDIA hardware.

• Extensive experience in Slurm workload manager and K8s.

• Understanding of parallel programming and experience with multi-threaded or distributed computing on NVIDIA platforms.