AI for Datadog & Datadog for AI (Live Demos)
Details
All talks will be presented in English, to ensure that as many people as possible can participate and engage at this event.
If you want to attend, please RSVP to secure your spot - this will make organizing easier. Thank you so much ♥️
Location: SVA System Vertrieb Alexander GmbH, Rotherstraße 16, 10245 Berlin, see travel information for more.
------
🏠 18:00 - Arrival: Networking, Drinks & Snacks (30 min.)
Grab yourself snacks & drinks and say hello to everybody else!
📅 18:30 - Introduction & What's new at Datadog? (15 min.)
🎙️Speaker: Marcel Drechsler, User Group Leader & Datadog Ambassador
Introduction into the evening and highlights of Datadog's recent new features and products.
📅 18:45 - Working Smarter With AI in Datadog: From Bits AI and Investigations to Full LLM Observability - Pilot to Production for your Projects! (40 min.)
🎙️Speaker: James Henderson, Partner Solutions Architect (EMEA) @ Datadog
Most of us use a fraction of what Datadog can actually do - and now there's powerful new capabilities: AI for Datadog and Datadog for AI. This session is a live demo: your chance to see the latest AI features in action and what they mean for your day-to-day work. We'll start with AI for Datadog - how Bits AI, Autonomous Investigations and natural-language querying take the toil out of troubleshooting, surfacing root cause in seconds. Then we'll flip it around to Datadog for AI: if you're shipping AI projects, LLM features or agents, we'll show you how AI LLM Observability traces every prompt, call and tool step, catches quality and cost drift early, and takes your AI from pilot to production with confidence. We'll spend fifteen minutes aligning on the new Datadog AI capabilities, then move straight into the platform for live demos showing what's possible.
📅 19:40 -The AI Agent Passed Every Test. Then It Shipped.
(30 min.)
🎙️Speaker: David Edoh-Bedi, Developer Relations @ Datadog
Your CI is green. Your evals passed in dev. You shipped the agent. Now what?
Production AI agents fail in ways your test suite was never built to catch. Prompts drift silently as people "optimize" them. Agents enter tool loops and burn through your API budget overnight. Your dashboards stay green and your users notice anyway.
This talk walks through real production failure modes using a live agent, shows what each one looked like from the outside (spoiler: fine), and demonstrates the monitoring that actually caught them.
You'll leave with specific instrumentation to add before your next agent ships, and a clearer sense of what your dashboards aren't telling you.
🥗 20:15 - Drinks, Food & Networking
Enjoy refreshments while networking with community peers!
👋 21:00 - Goodbye, see you next time!
