Multi-cloud AI infrastructure in action — with Nebius and dstack

Join us to explore how modern ML teams run large-scale training across multiple clouds — seamlessly and cost-efficiently.

This session shows how dstack orchestrates GPU workloads on Nebius infrastructure to accelerate experimentation, control spend, and stay vendor-neutral.

You’ll learn how to:

• Understand why GPU container orchestration matters — and what’s beyond Kubernetes and Slurm.
• Deploy your first distributed training job on Nebius (live walkthrough).
• Apply cost-optimization strategies for efficient GPU utilization.

Who should attend:
ML/AI Engineers, MLOps Engineers, Solution Architects, CTOs and anyone building or managing AI infrastructure at scale.