The Future of Inference

Hosted by The General Partnership ( https://www.thegp.com/ ), this private dinner will bring together a small group of leaders shaping the next phase of the inference landscape. The room is being curated across inference infrastructure providers, companies running inference at scale, and application teams whose products depend on high-volume token flow. The focus will be on how production inference is evolving as usage scales, including serving architecture, KV caching, disaggregation, latency, reliability, and cost/performance trade-offs. The goal is to create a candid, operator-level conversation around where the inference stack is heading, what is breaking in production today, and how leading teams are thinking about the technical and economic decisions that will define the next wave of AI applications.