Senior Solutions Architect, AI Performance Engineering

What you'll do

• Work with customer engineers to identify performance bottlenecks and develop parallel algorithms, data structures, and GPU solutions.
• Optimize operators through GPU kernel, instruction-level, and compiler tuning, and improve distributed communication for transformer workloads.
• Collaborate with NVIDIA architecture, research, libraries, tools, and system software teams to influence future platforms and programming models.

What they ask for

• 8+ years of hands-on, validated ML/DL performance engineering focused on GPU compute efficiency for training and inference.
• Experience with C, C++, or Python; proficiency with Linux.
• Solid understanding of software development, programming techniques, algorithms, linear algebra, and numerical methods.
• Background in parallel programming and high-performance computing, including parallel architectures and performance analysis and tuning.

Nice to have

• CUDA kernel development and experience with Nsight System and Nsight Compute.
• Distributed communication optimization; full-stack LLM or HPC evaluation and optimization experience.

Pay

184,000 USD - 287,500 USD base salary (Level 4); 224,000 USD - 356,500 USD base salary (Level 5)

Tools

C, C++, Python, Linux, CUDA, cuDNN, cuBLAS, CUTLASS

AI focus

Accelerates Physical AI workloads for autonomous vehicles and robotics, including training and inference performance optimization and distributed transformer communication.