End-to-end stack for building custom LLMs

In this webinar, we’ll walk through a modern stack for developing domain-specific LLMs using the latest post-training techniques — including custom reward functions, as popularized by Deepseek.

These cutting-edge methods are now within reach for any company, thanks to cost-efficient infrastructure from Nebius and a user-friendly software stack powered by Metaflow.

We’ll also showcase how the new Nebius — Outerbounds integration streamlines the end-to-end workflow for real-world production environments.

What you will know?

  • An introduction to sophisticated post-training, leading to reasoning models

  • Developing such custom LLMs in a production-grade environment

  • The components of an end-to-end stack for custom LLMs — what are they and why they matter

For who?

ML/AI engineers, Data Scientists, Research engineers, Platform engineers.

When?

May 20, Tuesday.

19:00 CEST (UTC +2) / 10:00 PDT (UTC –7)

Eddie Mattia

Data Scientist at Outerbounds

Alex Salikov

Head of External Solutions Department

Try Nebius AI console today

Get immediate access to NVIDIA® GPUs, along with CPU resources, storage and additional services through our user-friendly self-service console.