TensorRT Performance Engineer for Gen AI Inference

NVIDIA Corporation; Santa Clara (CA); On-site; USD 124,000 - 241,500; Full time

Job summary

NVIDIA Corporation seeks an experienced Deep Learning Software Engineer, TensorRT Performance, to analyze and enhance the performance of NVIDIA’s inference ecosystem, including TensorRT, TensorRT‑EdgeLLM, and Torch‑TensorRT.

You will establish benchmarking workflows, optimize model pipelines with quantization and scheduling, and collaborate with teams across AI, automotive, robotics and vision domains to deliver high-performance inference solutions.

Qualifications

• Bachelor’s, Master’s, PhD, or equivalent experience in CS/CE/AI.
• At least 2 years of relevant software development experience.
• Strong C++ and Python programming and software engineering skills.
• Experience with deep learning frameworks and inference libraries.
• Experience with performance analysis and optimization.
• Strong foundation of GPUs and modern DL workloads.

Responsibilities

• Establish groundbreaking performance benchmarking methodologies and analysis workflows and identify performance issues and opportunities for NVIDIA’s inference ecosystem.
• Contribute features and code to NVIDIA/OSS inference frameworks including TensorRT, TensorRT‑EdgeLLM, Torch‑TensorRT.
• Develop new model pipelines with optimized performance including quantization, scheduling, memory management, and distributed inference.
• Collaborate across generative AI, automotive, robotics, image understanding, and speech understanding teams to develop inference solutions.
• Scale performance of DL models across architectures and accelerators.

Skills

C++ programming

Python programming

Performance analysis

GPU architecture

DL model workloads

DL frameworks (PyTorch, JAX, Tensor?f)

Inference libraries (TensorRT)

Low-latency/edge AI

Graph compilers (TorchDynamo)

Embedded AI pipelines

Education

Bachelor's/Master's/PhD or equivalent

Tools

CUDA

TileIR

CuTeDSL

cutlass

Triton

TorchDynamo

TorchInductor

vLLM

TorchTensorRT

TensorRT

Job description

NVIDIA Corporation seeks an experienced Deep Learning Software Engineer, TensorRT Performance, to analyze and enhance the performance of NVIDIA’s inference ecosystem, including TensorRT, TensorRT‑EdgeLLM, and Torch‑TensorRT.

You will establish benchmarking workflows, optimize model pipelines with quantization and scheduling, and collaborate with teams across AI, automotive, robotics and vision domains to deliver high-performance inference solutions.