Event-Driven Workflows for Everyone: Orchestrating Python with Airflow

Modern systems don’t just run on fixed schedules, they need to react to events as they happen. From real-time API calls to streaming data or simple file drops, today’s workflows must be dynamic, reliable, and collaborative. Apache Airflow, long known as the go-to tool for batch scheduling, has evolved to handle this challenge through datasets, triggers, and event-driven orchestration.

In this talk, we’ll explore how to build event-driven workflows in Airflow that respond in real-time to data availability, messages from systems like Kafka, or signals from other applications. We’ll also look at patterns for hybrid orchestration (combining batch + streaming), and how to make DAGs more collaborative and self-service friendly so that analysts, scientists, and operations teams can engage with pipelines without needing to be Airflow experts.

The audience for this talk is Python developers, data engineers, and data scientists who are already familiar with building workflows but want to take them further into the event-driven world. Attendees will walk away with practical examples, design patterns, and a clearer understanding of how to make Airflow not just a scheduler for engineers, but a shared platform for entire teams to orchestrate data and processes in real-time.