Bauplan + dltHub: Python Based Data Infra for AI Agents

​AI agents are quickly becoming part of every software workflow. They write code, debug systems, and automate complex tasks. Their impact grows dramatically when they can interact directly with the data that powers a business.

​The next generation of infrastructure will be the one that enables agents to operate safely on real data.

​In this session, we'll show what that looks like in practice. Using Python, dltHub, and Bauplan, we'll walk through an end-to-end workflow where agents discover external data sources, generate ingestion pipelines, transform datasets, inspect results, and publish validated updates to production.

​The result is a stack that lets agents move beyond isolated tasks and start operating directly on the systems that drive real business decisions.