Delivering the Blueprint for Premium Inference: From Faster Tokens to Disaggregated AI Infrastructure
Session Overview
AI is moving from experimentation to production, and inference is becoming the real bottleneck. As enterprises adopt coding assistants, multimodal AI, and agentic workflows, success depends not only on model quality, but on delivering fast, reliable, secure, and cost-efficient intelligence at scale.
In this talk, Reggie will introduce premium inference as the next frontier of AI infrastructure. He will explain why latency, throughput, concurrency, memory efficiency, and cost per useful token matter more than ever. The session will also explore disaggregated inference as a new blueprint for scaling enterprise AI by separating prefill, decode, and orchestration across the right infrastructure layers.
This talk is designed for anyone who wants to understand how the next generation of AI infrastructure will power real-world, production-grade enterprise AI.
Speaker
Reggie Lu is an AI infrastructure and GenAI specialist at SambaNova, based in Tokyo. He specializes in helping enterprises design, deploy, and scale production-grade AI systems, with a focus on LLM inference, enterprise RAG, agentic AI applications, and secure on-premises AI infrastructure. He is particularly skilled in end-to-end AI productization β from customer requirements and architecture design to model serving, benchmarking, optimized deployment, and production operations. Previously, Reggie worked as an ML and AI infra software engineer, building AI accelerator software stacks, MLOps pipelines, cloud-native systems, AI applications, and enterprise platforms across multiple industries.