Summary
Kyle Kranen is a senior engineering leader at NVIDIA with a decade of experience building and scaling production deep learning systems, currently driving planetary-scale inference through the Dynamo project. He blends hands-on expertise in LLMs, distributed training and inference, and efficient model optimizations (quantization, speculation, config search) with people leadership across high-impact teams. Kyle led inference for NVIDIA AI Foundations and architected large-scale GNN training and deployment tooling capable of billion-node graphs, showing rare breadth across model families from VLMs to GNNs and time-series systems. He’s comfortable shipping low-level performance wins (TensorRT, VLLM, FP8/NVFP4) and high-level orchestration for datacenter-scale deployments. Based in San Francisco and rooted in a UC Berkeley EECS background, he pairs applied research instincts from academic medical imaging work with pragmatic engineering that accelerates real customer workflows. Beyond core technical depth, Kyle focuses on deep learning HCI—making powerful models accessible and usable outside narrow LLM interfaces.
10 years of coding experience
5 years of employment as a software developer
Bachelor’s Degree, Electrical Engineering and Computer Science, Bachelor’s Degree, Electrical Engineering and Computer Science at University of California, Berkeley
Menlo Atherton High School