Skip to content

Details

Registration on Meetup is invalid. Please register on Luma: https://luma.com/observability?utm_source=meetup
​When an AI application slows down or fails, the real cause can be buried across model calls, agent steps, tools, infrastructure and the database.
​This Dublin AI Week build session is about making those systems observable. We will instrument an AI application, connect useful signals and work through practical ways to find failures before they become production incidents.
​Expect two focused talks, a guided build session, food and time to meet other engineers and founders working on production AI systems.
​What we will build

  • ​Add practical observability to an AI application
  • ​Trace model, agent and tool activity across a workflow
  • ​Connect application signals with database metrics
  • ​Investigate a realistic production issue and identify its root cause
  • ​Leave with patterns you can apply to your own AI stack

​Talks
​AI Observability in Action: MongoDB Atlas OTel Metrics Sink
​15-minute talk and live demo by Harshit Mehta from MongoDB
​AI applications can fail or slow down because database issues are hidden from application telemetry. In this live demo, we’ll use the MongoDB Atlas OTel Metrics Sink to connect Atlas metrics with application signals and show how observability helps identify the root cause of a production AI issue.
​AI Observability by Bronto
​Bronto will share practical approaches to observing AI systems and understanding what is happening across production workflows before we move into the build session.
​Agenda

  • ​6:00 PM: Doors open, food and networking
  • ​6:30 PM: Welcome and build-session briefing
  • ​6:40 PM: MongoDB talk and live demo: AI Observability in Action: MongoDB Atlas OTel Metrics Sink
  • ​6:55 PM: AI Observability by Bronto
  • ​7:15 PM: Guided build session
  • ​8:30 PM: Builds, lessons and audience questions
  • ​8:45 PM: Food and networking
  • ​9:00 PM: Close

​Bring a laptop. Setup guidance and any technical prerequisites will be shared with approved guests before the event.
​Who it is for
​Software engineers, AI and ML engineers, technical founders, platform engineers and builders working on AI applications, agents or production infrastructure. You should be comfortable working through a technical build session, but you do not need prior observability experience.
​Partners
​**Bronto** is the headline partner for Dublin AI Week and is partnering on this practical session around AI observability.
​**MongoDB** is hosting the event at its Dublin office and contributing the live Atlas OTel Metrics Sink demo.
​We may take photos and short video clips during the evening for event recaps and community updates. If you prefer not to appear, tell one of the hosts when you arrive.
​Venue
​MongoDB Dublin
Building 2, Number 1 Ballsbridge
Shelbourne Road, Ballsbridge, D04 Y3X9, Dublin
​Dublin AI Week
​This event is part of Dublin AI Week, a citywide programme of practical AI events by Give(a)Go and Baseline. Explore the full programme.

Related topics

You may also like