Peter Kisfaludi

Peter Kisfaludi is a senior software engineer working in the TensorRT Multi-Device team. In this role, he focuses on developing scalable runtime architectures and optimizing communication overhead to deliver low-latency execution for multi-GPU model serving. Before joining NVIDIA in 2022, Peter was an independent consultant designing low-latency, mission critical software for real-time embedded systems.
Avatar photo

Posts by Peter Kisfaludi

Agentic AI / Generative AI Sep 21, 2026

Simplifying Model Serving Across Multiple GPUs with NVIDIA TensorRT Multi-Device Integration in NVIDIA Dynamo-Triton

The compute and memory demands of generative AI increasingly exceed what a single GPU can provide. NVIDIA TensorRT multi-device inference is a new capability... 7 MIN READ
Developer Tools & Techniques Jun 25, 2026

Scaling AI Inference Across Multiple GPUs Using NVIDIA TensorRT with Multi-Device Inference Support

Generative AI workloads are rapidly outgrowing the memory and compute budget of single GPUs. For inference developers building media generation pipelines, the... 11 MIN READ