Keynote

Datadog has always been driven by a broader vision of helping teams understand and operate complex systems. In this session, you’ll hear from Yrieix Garnier, VP of Product, and Hugo Kaczmarek, Senior Director of Product, as they share the latest updates across the Datadog product suite and discuss how that vision continues to shape the platform’s evolution and support the next generation of AI-driven applications.

From expanded agentic capabilities and new security and compliance features for the UK to improvements in data reliability for AI applications, join us for a first look at how Datadog helps teams understand and operate complex, distributed environments.

Virgin Atlantic’s brand is built on human warmth and wit. Can you build an LLM-powered AI experience that lives up to that promise? This was the question the Digital Engineering team at Virgin Atlantic had when they started building Concierge, their AI travel-savvy sidekick.

In this session, Mark O’Neill, Senior Manager of AI Software Engineering, shares how they built Concierge from scratch and evolved it safely in production, where hallucinations and unexpected responses pose real brand and safety risks. He will cover how Red and Blue teaming techniques were used to test behavior and improve response quality, and how Datadog LLM Observability became a critical tool for both engineering and product teams. From debugging real-world issues to learning from user interactions, observability helped shape decisions and guide the product roadmap.

Attendees will leave with concrete patterns for testing LLM behavior, operating AI systems in production, and using observability data to make product decisions... all without compromising brand trust.

In modern sports trading, milliseconds matter. At Entain, real-time odds are continuously recalculated based on live game state, player performance, weather conditions, and dozens of other signals. Behind the scenes, this means running ultra-low-latency, distributed pricing pipelines where mathematical models can execute hundreds of thousands of simulations and still deliver updated prices end-to-end in under 400 milliseconds.

In this session, Adam Hope, Global Trading CTO at Entain, takes you inside the systems that power live sports trading at scale. He’ll walk through the architecture behind Entain’s real-time pricing platform, from data streaming and distributed services to latency-sensitive model execution. He’ll cover how, using Datadog APM, data streaming observability, and SLOs, they are able to continuously validate correctness and latency across critical services, ensuring odds are recalculated accurately and delivered on time across global markets operating 24/7.

Attendees will learn how instrumenting critical paths with APM and data streaming observability in their latency-sensitive systems, can help them validate those systems end-to-end, detect degradation early, and ensure reliability as systems and load scale.