Nebius.Build/BER is a technical in-person event for engineers, architects, founders, and developers building, deploying, and scaling AI systems in production.

The program features talks from Nebius and partners including Anyscale and Tavily, with practical insights into distributed training, running open-source models in production, inference scaling, and operating reliable AI workloads at scale. It also includes a hands-on workshop where participants can deploy their own model and connect it to a working inference pipeline.

Why attend?

Expect technical sessions instead of marketing, real architecture learnings from production environments, and direct exchange with engineers on site.

Agenda

12:00–13:00 Registration and lunch

13:00–13:15 Opening remarks

13:15–13:45 Talk: Reliable clusters for large-scale distributed training

13:45–14:30 Talk: Making open models reliable at scale

14:30–15:00 Coffee break

15:00–15:25 Breakout sessions: Anyscale talk or OpenClaw workshop

15:25–15:50 Agentic Search / workshop continuation

15:50–16:05 Keynote

16:05–16:20 Fireside chat

16:20–16:30 Closing remarks

From 16:30 Networking and help bar

This event is ideal for ML infrastructure engineers, AI/ML platform engineers, architects, technical founders, and developers working on scalable AI systems.

Register here:

https://nebius.com/events/nebius-build-berlin

DISCLAIMER:

This is a third-party event announcement. We are not responsible for it and are only promoting it. If you would also like to feature your event here, please contact us.