Inference Performance Engineer: AI GPU Optimization&Equity

NVIDIA; Santa Clara (CA); Hybrid; USD 124,000 - 242,000; Full time

Benefits offered by this job

Equity

Benefits package

Job summary

NVIDIA is seeking a Senior Inference Performance Engineer to push performance limits on large-scale AI inference benchmarks. You will optimize AI model execution and profiling workflows, enabling autonomous optimization across TensorRT-LLM, vLLM, and disaggregated serving architectures.

The role emphasizes measurable gains, reproducible experiments, and close collaboration with GPU and software teams. Responsibilities include distilling performance methods, benchmarking with Nsight tools, and

Qualifications

• Must have extensive knowledge of AI model execution efficiency and optimization.
• Hands-on benchmarking and profiling GPU workloads with Nsight Systems, Nsight Compute, CUPTI, or PyTorch profiler.
• Strong Python and the ability to navigate/modify large C++/CUDA codebases.
• Rigorous experimental methodology with controlled single-variable comparisons and reproducible benchmarks.
• Excellent written and verbal communication of performance tradeoffs.

Responsibilities

• Distill performance instincts into reusable skills, workflows, and evidence-backed methodologies for autonomous AI agents.
• Improve throughput-per-GPU and interactivity via configuration options, batching, and memory management.
• Profile workloads using Nsight tools and analyze with roofline/latency metrics to drive fixes.
• Land upstream improvements: serving framework patches, optimized kernels, and deployment recipes.
• Collaborate with TensorRT-LLM, SGLang, vLLM, kernel, benchmarking, and GPU teams to convert profiling insights into gains.
• Communicate tradeoffs clearly to humans and documentation for autonomous systems.

Skills

Python

C++/CUDA

Performance optimization

Experimental methodology

Education

BS/MS/PhD in CS/CE/EE or related field

Tools

Nsight Systems

Nsight Compute

CUPTI

PyTorch profiler

TensorRT-LLM

Job description

NVIDIA is seeking a Senior Inference Performance Engineer to push performance limits on large-scale AI inference benchmarks. You will optimize AI model execution and profiling workflows, enabling autonomous optimization across TensorRT-LLM, vLLM, and disaggregated serving architectures.

The role emphasizes measurable gains, reproducible experiments, and close collaboration with GPU and software teams. Responsibilities include distilling performance methods, benchmarking with Nsight tools, and