The Future of Inference & Compute

​The Future of Inference and Compute
A private dinner hosted by Nebius and Timeless

​An intimate evening in San Francisco for founders, researchers, and infra engineers thinking hard about what it takes to serve AI at scale.

​Inference is on track to become one of the largest computing markets in history, and the bottlenecks are physical: memory bandwidth, power, interconnect, and the real cost of every token. Over dinner we’ll get into where the stack is heading: when it makes sense to build your own inference stack versus use a managed one, and how post-training and serving are converging into a single applied-AI problem.

​Featuring Ryan Hanrui Wang, SVP, Applied AI at Nebius

​Ryan leads inference, post-training, and agentic systems at Nebius, including Nebius Token Factory. He co-founded and was CEO of Eigen AI, acquired by Nebius in 2026, whose inference platform was named the #1 speed inference provider in Jensen Huang’s GTC 2026 keynote and ranked #1 on 25+ models on Artificial Analysis. He earned his Ph.D. at MIT and was named to MIT Technology Review’s Innovators Under 35.

​About Timeless:

​

​About Nebius:

​Nebius is the AI infrastructure platform backed by NVIDIA, purpose-built for teams training, fine-tuning, and deploying at scale.

​__________________________________________________

For information on how Nebius processes the personal information you provide when registering, please see Nebius Privacy Policy and Event Privacy Policy. Nebius and partners are committed to managing your personal information carefully and in compliance with (UK) GDPR and further applicable data protection legislation.