Eric Phan
System Software Engineer at NVIDIA
Eric Phan is a System Software Engineer at NVIDIA who specializes in a diverse array of AI products, including data science, robotics, and computer vision. He focuses on enhancing the developer experience across NVIDIA’s AI offerings. Prior to joining NVIDIA, he conducted AI research on UAVs/ASVs at Georgia Tech Research Institute.
All Sessions by Eric Phan
Accelerated Data Science DLI X-Lab
<span class="etn-schedule-location"> <span class="firstfocus">LLMOps & MLOps</span> <span class="firstfocus">All Levels</span> </span>With datasets growing rapidly in volume, velocity, and veracity, the demand for more efficient data processing has never been higher. This workshop walks through how to apply open-source GPU accelerators from the NVIDIA RAPIDS project for common python data science workflows. In this training, you will: Speed up data manipulation tasks in pandas using cuDF, NVIDIA's GPU-accelerated DataFrame library, and explore advanced operations like grouping, sorting, and merging data. Discover how Polars enhances larger scale data processing with memory optimizations, lazy execution, and cuDF-powered GPU parallelization Explore GPU-accelerated feature engineering and ML modeling with XGBoost Perform graph analytics faster and more efficiently using nx-cugraph, a NetworkX backend that accelerates most of its popular algorithm
Accelerated Data Science DLI X-Lab
<span class="etn-schedule-location"> <span class="firstfocus">LLMOps & MLOps</span> <span class="firstfocus">All Levels</span> </span>With datasets growing rapidly in volume, velocity, and veracity, the demand for more efficient data processing has never been higher. This workshop walks through how to apply open-source GPU accelerators from the NVIDIA RAPIDS project for common python data science workflows. In this training, you will: Speed up data manipulation tasks in pandas using cuDF, NVIDIA's GPU-accelerated DataFrame library, and explore advanced operations like grouping, sorting, and merging data. Discover how Polars enhances larger scale data processing with memory optimizations, lazy execution, and cuDF-powered GPU parallelization Explore GPU-accelerated feature engineering and ML modeling with XGBoost Perform graph analytics faster and more efficiently using nx-cugraph, a NetworkX backend that accelerates most of its popular algorithm
