Steven Hillion

When a dashboard breaks, someone files a ticket. When a regulatory reporting deadline is missed, someone calls the regulator. When the AI models behind a travel site go stale, someone books somewhere else. Steven Hillion from Astronomer looks at the pipelines companies genuinely cannot run without, what it takes to keep them up, and how to avoid the silent failures that are quietly leaking revenue. Steven will share two real-world examples of data teams using Airflow and Snowflake. Arch Insurance orchestrates pipelines underpinning their regulatory and actuarial reporting. Booking.com trains and serves the models behind billions of searches a year and the trip-planning assistants customers interact with directly. In both cases, pipeline failures can be costly: one ends in a fine, the other in lost revenue. Building and maintaining resilient, mission-critical pipelines like these require both a proactive approach to reliability and scalability to avoid failures and robust observability, SLAs, and swift, automated recovery when something inevitably breaks.

Every enterprise has an AI strategy. Few have the foundation to deliver on it. Most treat context as something defined upfront and consumed downstream. Semantic layers define metrics. Ontologies model relationships. Neither captures what actually ran. Enterprise context is not static. It is evidence that accumulates continuously: which pipelines ran, what changed upstream, what humans corrected. Orchestration is the only layer where that evidence accumulates naturally, not as declared truth, but as a continuous record of what actually happened. That is what production AI runs on. This talk outlines the architectural pattern that makes it possible.

Most enterprises have an AI strategy, but few have the data foundation to deliver lasting value. Leading teams are differentiating themselves with systems that capture execution metadata from across the data estate. In conversation with Geodis, we’ll ground the conversation in real-world experience—sharing lessons learned and the critical investments required to move from planning to production-grade AI with measurable results.

The role of data teams and data engineers is evolving. No longer just pipeline builders or dashboard creators, today’s data teams must evolve to drive business strategy, enable automation, and scale with growing demands. Best practices seen in the software engineering world (Agile development, CI/CD, and Infrastructure-as-code) from the DevOps movement are gradually making their way into data engineering. We believe these changes have led to the rise of DataOps and a new wave of best practices that will transform the discipline of data engineering. But how do you transform a reactive team into a proactive force for innovation? We’ll explore the key principles for building a resilient, high-impact data team—from structuring for collaboration, testing, automation, to leveraging modern orchestration tools. Whether you’re leading a team or looking to future-proof your career, you’ll walk away with actionable insights on how to stay ahead in the rapidly changing data landscape.

Today's data teams are evolving beyond pipeline builders to become strategic business enablers. This fundamental shift requires embracing principles that transformed software engineering — agile methodologies, CI/CD practices and infrastructure-as-code — and are now making their way into data workflows as DataOps. This session explores essential strategies for building resilient, high-impact data operations. We'll dive into team structures that foster collaboration, testing frameworks that ensure reliability, automation approaches that eliminate toil, and modern orchestration tools that scale with growing demands.

Sensor data on the event stream can be voluminous. In NAND manufacturing, there are millions of columns of data that represent many measured and virtual metrics.