Nebius AI Cloud Unveiled. San Francisco Meetup
At Nebius, we’re creating an AI Cloud — a cutting-edge cloud platform designed exclusively for AI practitioners. We provide seamless AI solutions accelerated by NVIDIA, popular MLOps tools and a scalable inference platform to support AI application development.
Ready to meet the team behind it all?
Join us at Convene, San Francisco, on Thursday, 13 March, for an exclusive look into Nebius AI Neocloud, our cutting-edge platform designed for explorers and innovators.
Whether you’re developing AI models, scaling ML workloads, or deploying AI applications, this is a great opportunity to connect with our developers, product experts, architects and some of our partners.
What’s in store?
-
An insider’s look at Nebius architecture and development principles.
-
Insights into the NVIDIA Blackwell Architecture.
-
A deep dive into how we developed a Kubernetes operator for Slurm.
-
Discover how test-time computation unlocks the potential of agentic systems.
And much more!
Plus, all attendees will receive free credits to try out our GPU Cloud Accelerated by NVIDIA and explore the new text-to-image functionality in AI Studio.
Spaces are limited, so don’t miss out! Register now to secure your spot.
Agenda
1 pm — Registration & welcome coffee
Start the day with networking and refreshments.
2 pm — Introduction
Vasily Pantyukhin, Nebius
Setting the stage for an afternoon of innovation and discovery
2:15 pm — How we build cloud for AI workloads
Gleb Kholodov, Nebius
An in-depth look at the hardware and software architecture powering AI applications.
2:45 pm — Marrying Slurm and Kubernetes for Workload Management
Dmitry Starov, Nebius
Discover Soperator, a Slurm-based workload manager for ML and HPC clusters.
3:15 pm — Break
3:30 pm — What it takes to win the large-scale training game: we made the mistakes, so you don't have to
Levon Sarkisyan, Nebius
Avoid the top mistakes and master best practices for scaling AI models.
4 pm — TractoAI — a serverless platform for AI engineers (case study with SynthLabs)
Maxim Akhmedov & Mike Burkov, Nebius
Nathan Lile & Louis Castricato, SynthLabs
Customer spotlight.
4:45 pm — Improving agentic systems with test-time computation
Karina Zainullina, Nebius
Talk about our recent research on combining guided search with agent inference, and how these techniques enable us to build better software engineering agents.
5:15 pm — Inference, all you need to know about it
Alex Patrushev & Roman Gaev, Nebius
Our experience on building inference. Tips on choosing inference providers and GenAI models.
5:45 pm — Closing remarks
Vasily Pantyukhin, Nebius
6:00 pm — Networking, drinks and bites
Speakers
Convene
100 Stockton St, San Francisco, CA 94108, USA
Please make sure to bring your confirmation email
Register now
Please register to express your interest in attending.
You’ll receive an email to confirm if your application has been successful.