Inference Performance Engineer - Agent-Driven Optimization

NVIDIA Corporation; Santa Clara (CA); On-site; USD 124,000 - 196,000; Full time

Job summary

NVIDIA Corporation seeks a Senior Inference Performance Engineer to push the performance limits of large-scale AI inference benchmarks. You will optimize autonomous agent workflows, profiling, and tuning processes to maximize GPU throughput and user interactivity.

Responsibilities include collaborating across teams to roll out patches, optimized kernels, and deployment recipes while maintaining model correctness.

Qualifications

• Proven expertise in optimizing AI model execution for throughput and latency.
• Experience benchmarking and profiling GPU workloads with established tools.
• Strong Python engineering skills; able to modify large C++/CUDA codebases.
• Rigorous experimental methodology and reproducible benchmarking practices.

Responsibilities

• Distill performance insights into reusable workflows for autonomous AI agents.
• Improve AI inference throughput and interactivity through configuration, batching, and memory optimizations.
• Profile and optimize serving architectures across TensorRT-LLM, vLLM, and related tools.
• Collaborate with cross-functional teams to implement patches and deployment recipes.

Skills

Python

C++/CUDA

Education

BS/MS/PhD in CS/CE/EE/Applied Math

Tools

Nsight Systems

Nsight Compute

CUPTI

PyTorch profiler

Job description

NVIDIA Corporation seeks a Senior Inference Performance Engineer to push the performance limits of large-scale AI inference benchmarks. You will optimize autonomous agent workflows, profiling, and tuning processes to maximize GPU throughput and user interactivity.

Responsibilities include collaborating across teams to roll out patches, optimized kernels, and deployment recipes while maintaining model correctness.