Thursday, June 11, 2026 | 5:30 PM (GMT+10)
Stone & Chalk @Tech Central, Level 1, 477 Pitt St Haymarket NSWJoin us this June in Sydney for an engaging meetup focused on building high-performance data infrastructure for the AI era. Featuring Apache Doris PMC Chair and Principal Solution Architect at VeloDB, this event will dive into the necessity of real-time analytics for AI agents and offer practical strategies for AI observability through a live technical demonstration.
| Time | Agenda |
|---|---|
| 5:30 PM | Doors Open |
| 6:00 PM | Welcome! |
| 6:10 PM | When Every Second Counts: Real-Time Analytics for AI Systems Mingyu Chen, Apache Doris PMC Chair & VPE@VeloDB AI applications are changing what “real-time” means. It is no longer enough to analyze data after the fact. In AI-driven products, decisions often need to happen while events are still unfolding: a user is interacting with an agent, a model is generating output, a workflow is waiting on a tool call, or an operational issue is emerging in production. This talk introduces real-time analytics and explains why it has become more important in the AI era. We will look at common AI-era use cases, including observability, user behavior analysis, operational monitoring, and agent performance tracking. Using Apache Doris as the implementation layer, we will explore how a real-time analytics system can support fast ingestion, low-latency queries, and scalable interactive analysis. I will also share a brief introduction to the Doris open source community and demonstrate a real-time observability use case for AI agents. The goal is to show how teams can turn streaming events into actionable insights quickly enough to improve user experience, reliability, and decision-making. |
| 6:40 PM | Realtime AI observability and analytics with Apache Doris - Live Demo Shilin Wu, Principal Solution Architect@VeloDB AI systems don’t behave like traditional software. When something goes wrong, it’s rarely obvious why. In this live session, we’ll walk through how you can start making AI systems more observable in practice. We’ll explore how to: 1. Understand what your AI system is actually doing under the hood 2. Trace how outputs are generated across different steps 3. Identify where things start to drift or break down 4. Analyze usage patterns, performance, and cost over time We’ll also show how Apache Doris can be used to store, query, and analyze AI interaction data at scale—so you can move beyond intuition and actually reason about your system. This is a live, evolving demo focused on ideas you can take back and apply immediately. |
| 7:10 PM | Networking |
| 7:30 PM | Doors Close |
How to find us
Stone & Chalk @Tech Central, Level 1, 477 Pitt Street Haymarket NSW
Head up to Level 1, Stone & Chalk and we will be in the workshop space on the left as you come through the doors
Speakers
Mingyu Chen
Apache Doris PMC Chair & VPE@VeloDB
Mingyu Chen is the Apache Doris PMC Chair and Vice President of Technology at VeloDB. As a seasoned big data developer and one of the founding engineers of the Apache Doris project, Rayner has been a pivotal force in driving the community's growth and technical roadmap. Throughout the project's history, he has spearheaded the technical innovations across every major version and architectural evolution, nurturing Apache Doris into the flourishing analytical database it is today.
Connect with Mingyu on Linkedin
Shilin Wu
Principal Solution Architect@VeloDB
Shilin is Principal Solution Architect at VeloDB, specializing in the design and optimization of distributed real-time analytical platforms. With an impressive career spanning industry leaders such as Google, Ant Financial, and Confluent, Shilin brings a wealth of expertise in architecting high-performance data systems at scale. He is dedicated to helping organizations successfully transition Apache Doris and VeloDB into production, ensuring their environments are robust, scalable, and highly efficient.
