Senior Software Engineer, Deep Learning Inference - Tensorrt

NVIDIA NVIDIA · Semiconductors · Santa Clara, CA

NVIDIA is seeking a Senior Software Engineer to develop and scale a state-of-the-art inference framework for accelerating Deep Learning models, particularly LLMs, on NVIDIA GPUs using TensorRT. The role involves crafting inferencing software, developing components of TensorRT, and optimizing the deployment of trained models using C++ and Python.

What you'd actually do

  1. Craft and develop robust inferencing software that can be scaled to multiple platforms for functionality and performance
  2. Develop components of TensorRT, NVIDIA’s SDK for high-performance deep learning inference.
  3. Closely follow academic developments in the field of artificial intelligence and feature update TensorRT
  4. Use C++ and Python to build graph parsers, optimizers, and tools for effective deployment of trained deep learning models.
  5. Collaborate with teams of deep learning experts, GPU architects and DevOps engineers across diverse teams.

Skills

Required

  • Bachelor's, Master's, PhD or equivalent experience in Computer Science, Computer Engineering, Electrical Engineering or related field
  • 3+ years of software development experience
  • Strong experience with the latest C++ standardsC++11/C++14/C++17/C++20, etc.
  • Strong grasp of Machine Learning concepts
  • Experience and knowledge in Computer Architecture, Data Structures, Algorithms
  • Excellent communication skills, and an aptitude for collaboration and teamwork

Nice to have

  • Experience developing System Software
  • Proficiency in Python
  • Background in GPU kernel programming using CUDA or OpenCL
  • Experience in software performance benchmarking, profiling, and optimizations
  • Background in compiler development
  • Experience in working with TensorRT, PyTorch, TensorFlow, ONNX Runtime or other ML frameworks

What the JD emphasized

  • high-performance deep learning inference
  • effective deployment of trained deep learning models

Other signals

  • inference framework
  • accelerating Deep Learning models
  • Large Language Models
  • NVIDIA GPUs
  • high-performance deep learning inference
  • deployment of trained deep learning models