Pavel Shamis

Pavel (Pasha) Shamis is a distinguished engineer at NVIDIA in the AI Data-Infra Optimization group where his primary focus lies in optimizing efficiency of the AI software and hardware stack. Before joining NVIDIA, Pasha served as a senior principal research engineer at Arm for six years, working on co-designing software and hardware building blocks for large-scale distributed systems.
Avatar photo

Posts by Pavel Shamis

Networking / Communications Dec 10, 2025

Enhancing Communication Observability of AI Workloads with NCCL Inspector

When using the NVIDIA Collective Communication Library (NCCL) to run a deep learning training or inference workload that uses collective operations (such as... 6 MIN READ
Image shows cloud-based GPU clusters dedicated to AI training.
Data Center / Cloud Mar 10, 2025

Ensuring Reliable Model Training on NVIDIA DGX Cloud

Training AI models on massive GPU clusters presents significant challenges for model builders. Because manual intervention becomes impractical as job scale... 8 MIN READ