From building agentic data platforms to solving massive-scale performance challenges, this session combines technical insights and real-world experiences from leading teams like Pinterest and Uber. Whether you’re a data engineer, infrastructure builder, or AI practitioner, you’ll walk away with hands-on perspectives and open-source insights to design the next generation of data systems.

​🍕 Drinks and food will be provided, along with plenty of time for networking and discussions with fellow builders.

​

​Agenda & Topics:

• ​​5:30 PM – 6:00 PM: Registration & Networking
• ​​6:00 PM – 7:45 PM: Opening Remarks & Speaker Presentations
• ​​7:45 PM – 8:30 PM: Open Networking & Conversations
• ​Topic 1: Re-Architecting Metadata for the Future – Lessons from Uber
  Jack Song – Head of Data Platforms & AI, Engineering Director at Uber
  → Discover how Uber re-architected its metadata systems to manage 200B+ entries, enabling dual API support, reducing operational complexity, and preparing for future growth.
• ​Topic 2:Taming Data at Scale: Pinterest's Journey to High-Performance AI with Iceberg and Parquet
  Ashish Singh – Principal Engineer, Data Engineering at Pinterest
  → This talk will share how Pinterest solved key data performance challenges using Iceberg and Parquet — particularly around schema evolution, fast backfills, data exploration, joins, and cost efficiency.
• ​Topic 3: Bridging Data & AI: How AI Agents Transform Data Management
  Jerry Shao -Co-founder and CTO of Datastrato
  → Explore how metadata-driven architectures (unified policies, job orchestration, statistics) enable autonomous and context-aware operations, moving organizations from manual processes to intelligent, agent-driven workflows.
• ​Topic 4: Building the Real-Time Lakehouse on AWS: A Unified Data Foundation for AI
  Yuankai (Kevin) Shen - Principal Product Manager at VeloDB
  → The Real-Time Lakehouse combines the scalability of data lakes with the speed of real-time analytics. This session will demonstrate how AWS Glue, S3 Tables, and Apache Doris work together on AWS to power low-latency, AI-ready insights. A live demo will highlight how this unified stack supports intelligent query performance and next-gen data workloads.

​Why Join?
✔ Get a first-hand look at cutting-edge approaches to AI-native data infrastructure
✔ Learn from real-world case studies at global scale
✔ Connect with peers in the data & AI community over food and drinks

​Host together？

​If you have suggestions for future discussion topics or are interested in sharing your own experiences at upcoming meetups, feel free to reach out—I’d love to hear from you!

​👉 Connect with me on LinkedIn Yue Guo