Melbourne Data Eng meetup, May Edition

Hey all, please join us for our May Edition of the Melbourne Data Eng meetup. Join us for an engaging session filled with exciting discussions and networking opportunities.

Don't miss out—RSVP to secure your spot. Please reach out if you'd like to host us! 🎉

🏠 Location: Stone & Chalk Melbourne
🍕 Catering: Cloud Shuttle 😊
💬 Join our Slack Group here: Data Engineers Slack Channel

Schedule:
- 5.30pm: Doors Open
- 6.00pm: Announcements and Welcome
- 6.10pm: 1st Talk
- 6.40pm: 2nd Talk
- 7.10pm: Networking
- 7.30pm: Doors Close

Speakers:
🎤 Erfan Hesami, Senior Data Enigneer/Architect @ Airmaster
Talk: Building an AI-Powered Data Catalouge with OpenMetadata
Summary: Ever ask an AI tool a question about your data and get a confident but wrong answer? You’re not alone. This usually happens because the AI is missing context.
LLMs are only as good as the information you give them. Without a strong metadata layer, even the best models don’t really understand your data. In this session, we’ll show how you can fix that.
I’ll walk through how OpenMetadata, MCP (Model Context Protocol), and its AI SDK work together to give AI the context it needs to understand your data, not just query it. Through a live demo based on a real project, you’ll see how to connect an AI agent to your metadata so it can search tables, follow data lineage, and answer questions with real understanding.
We’ll also show how the AI SDK helps you go beyond chat, building agents and workflows that can take action on your data, not just talk about it.
Bio: I am a Senior Data Engineer/Architect. I started my career as a Data Analyst, then moved into Analytics Engineering, and now focuses on building data platforms and systems.
I have worked with companies such as Allianz, Xero, AFL, and Airmaster, helping teams better use their data.
I enjoy sharing what I learn about data engineering and AI through simple, practical content on my Substack, Pipeline to Insights, helping others grow in their careers.

🎤 Peter Hanssens, Founder & Principal AI Enablement Engineer @ Cloud Shuttle
Talk: The AI Gateway: Your Organisation's Most Underrated Data Source
Summary: Every AI request your organisation makes generates a structured event: timestamp, user ID, model, tokens, cost, latency, cache hit, tool calls.
Most teams are discarding all of it. This talk is about building the data engineering layer around your AI Gateway — turning it from a routing proxy into a competitive analytics asset.
I'll cover the schema you get for free, the five DE problems you have to solve (schema normalisation, pricing enrichment, prompt/response storage tiers, org attribution, volume at scale), the dimensional model that makes it queryable, and what you can answer once the pipeline runs.
Three deep dives: prompt version A/B testing at production scale, cache effectiveness analytics, and agent tool call observability.
Bio: Peter is the founder of Cloud Shuttle, a Sydney-based data and AI engineering consultancy, and the founder of DataEngBytes - ANZ's largest data engineering community conference. He works at the intersection of data engineering and production AI: LLMOps, AI gateways, evaluation pipelines, and the infrastructure that makes AI measurable.

If you'd like to present at future meetups, please submit your talk here!

Remember to bring along some great questions! See you all there...