AI Computing Software Development Engineer, TensorRT-LLM
Core
Develop robust, scalable inference software and optimize performance for Large Language Models (LLM) across multiple platforms.
Role type
Senior IC software development engineer (LLM inference)
Builds
TensorRT-LLM inference software and runtime features
Domain
AI / Deep Learning / GPU Computing
Deliverable
production ML models
Required skills
Python, C/C++, software design, LLM inference techniques, deep learning frameworks (PyTorch, HuggingFace), performance analysis, kernel implementation
Preferred skills
LLM inference frameworks (TensorRT-LLM, SGLang, vLLM), DL compilers, performance modeling, profiling, debugging, code optimization, CPU/GPU architecture, GPU programming (CUDA, OpenCL)
Technologies
TensorRT-LLM, PyTorch, HuggingFace, CUDA, OpenCL, C/C++, Python
Responsibilities
Craft and develop robust inference software scalable to multiple platforms; Perform performance analysis and optimization for LLM inference; Implement kernels and runtime features to support new LLM models and inference algorithms; Provide feedback into architecture and hardware design; Collaborate with software, research, and product teams to guide deep learning inference direction
Seniority
Senior, hands-on IC
