The Data Engine Behind AI: Powering Language Models at Scale

This talk was presented at the ClickHouse Melbourne Meetup on September 30 by Alexey Milovidov, Co-Founder & CTO of ClickHouse.

The Data Engine Behind AI: Powering Language Models at Scale

Behind every AI breakthrough is a mountain of data — and the real challenge is moving, shaping, and querying it fast enough to keep up. In this session, Alexey will take a deep dive into how large language models are trained, fine-tuned, and served, with a focus on the data pipelines and architectures that make it all work. You’ll see live demos of ClickHouse tackling massive AI-scale datasets in real time, along with performance benchmarks that reveal just how far you can push the engine. Expect actionable tips, architectural patterns, and a few ClickHouse tricks that data engineers, architects, and AI enthusiasts can bring straight into their own projects.