AI駆動データエンジニアリングの発表

AI coding agents are changing how data engineers work.Over the past year we have been thinking carefully about how to make Dagster the best framework to use alongside an AI agent. Today, we are introducing a new addition to Dagster University: AI-Driven Data Engineering.Jump to the AI-Driven Data Engineering course pageLearnTeams at many of the most exciting companies in the world use Dagster to build sophisticated data platforms. Our goal has always been to make those capabilities accessible to every user.AI-Driven Data Engineering is an eight-lesson course that takes you from a blank directory to a production-ready ELT pipeline built entirely from prompts.Along the way you will learn how to:Choose the right AI toolsGuide an agent toward Dagster best practicesDevelop judgment about when to trust generated output and when to verify itEach lesson builds on the last. By the end, you will have a complete project ready to deploy.Who This Course Is ForThis course is designed for two groups of engineers.1. Engineers who want to learn Dagster or become Dagster power usersIf you are new to Dagster, this course shows you how to build a real project from scratch using modern workflows. If you already use Dagster, it will help you learn patterns that experienced teams rely on to move faster and build reliable pipelines.2. Engineers exploring agentic coding workflowsMany teams are still figuring out how to best use AI to help them build effectively.This course focuses on the practical side of that shift: how to structure a project so an agent can contribute reliably, how to guide an agent toward good architecture, and how to manage sessions and context when working on complex systems.In other words, it teaches both Dagster best practices and how to get the most out of AI-assisted development.AI-Driven Development with DagsterTo support this new way of working, we have invested in tooling that makes AI-driven development reliable and practical:An opinionated CLI (dg) that gives agents deterministic, validated actionsMaintained agent skills that keep output aligned with Dagster best practicesComponents that provide predictable structure for common integrationsHistorically, becoming a Dagster power user required internalizing a set of patterns and best practices. That created a real gap between getting started and moving quickly with confidence. But AI agents change that.With the right tooling and structure, an agent can help guide users toward good patterns from the start. The goal is not to replace expertise, but to make that expertise more accessible.Our aim is simple: help every data engineer become a power user of Dagster.Enroll TodayLike all Dagster University courses, AI-Driven Data Engineering is free and open to everyone. Sign up for Dagster University to get started. Once enrolled, you can track your progress and complete the course at your own pace.Jump to the AI-Driven Data Engineering course page‍