Ruixiang W.
Senior Developer Technology Engineer - Llm & Genai @NVIDIA
Signup · Get unlimited contacts
WORK HISTORY
Senior Developer Technology Engineer - Llm & Genai @NVIDIA
DE
End-to-end performance optimization for LLM training and inference* Advanced techniques in speculative decoding and model quantization* Applied research and innovation in LLMs and generative AI
EDUCATION
RWTH Aachen University
Master's degree, Computer Science
Xidian University
Bachelor's degree, Software Engineering
ABOUT RUIXIANG W.
Senior Developer Technology Engineer at NVIDIA, focused on accelerating LLM and generative AI workloads at scale. I specialize in optimizing both training and inference performance on GPUs, leveraging cutting-edge techniques such as model quantization, speculative decoding, system-level profiling, CUDA/Triton kernels, advanced fine-tuning optimizations etc. I collaborate closely with industry partners to unblock performance bottlenecks, enabling partners to achieve SoL(speed of light) performance on NVIDIA GPU platforms leveraging NVIDIA technologies.Previously, I worked extensively on applying machine learning and NLP to search technologies, including fine-tuning LLMs for tasks such as query understanding, related query generation, and review summarization. My background also includes distributed training and inference, parameter-efficient fine-tuning, and embedding-based retrieval systems enhanced through contrastive learning.I bridge the gap between AI research and high-performance production systems.Tech Stack: Python, C++, PyTorch, Transformers, DeepSpeed, ONNX, TensorRT, vLLM, Trition/CUDA, Nsight system profiling, Apache Spark/Hadoop, Docker etc.(and of course Agentic AI and Vide Coding/Engineering).
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.