Seattle Apache Iceberg™ Community Meetup! 🧊❄️🍦
Join us on July 15th (Tuesday) from 5:00-8:30 PM at the Snowflake Bellevue Office.
Connect with fellow enthusiasts, share insights, and dive into the latest developments in the Apache Iceberg™ ecosystem! Whether you're a seasoned pro or new to Apache Iceberg, this meetup is the perfect place to exchange ideas and spark innovation.
Note on Parking: You are welcome to use our complimentary visitor parking in the Snowflake Garage located at 123rd Ave NE. Drive down the ramp and take a left at the bottom (Level A). Park in any available parking space. After parking, follow the signage to the garage elevator, up to Level 1. From there, proceed right to the Building Lobby.
Agenda
5:00p - 6:00p: Doors Open & Networking 💃
6:00p - 7:30p: Welcome Remarks & Presentations!
7:30p - 8:30p: More Networking 🕺
The event will focus on innovations in Apache Iceberg (https://iceberg.apache.org/)
We will discuss topics around Open-Source Data Analytics, Open Table Formats (OTF), software concepts like Transactional Data Lakes or Lakehouse, advancements in AI/ML including generative AI, and many more topics of mutual interest that leverage Apache Iceberg.
During the sessions, we will provide you tips to get involved within the community, you will learn more about how the community is collaborating to grow the technology, and software/solutions that ease problem solving and improve user experiences.
Topic 1: Working with Iceberg Predicates: Representation, Translation, and Interop
Abstract
We will discuss how Daft's rust-based optimizer is able to push expressions into pyiceberg for optimizing reads. We will start by covering Daft's expression representation, and what we need to do to make them readable by pyiceberg. It's trickier than you think! And it motivates us to integrate directly with iceberg-rust. Then we will cover the challenges of various expression representations, the attempts others have made to use a consistent representation, and how we might get both table formats like Iceberg and Deltalake to consistently integrate with engines and databases like Daft, DataFusion, and Lance.
Conner Howell (Software Engineer @ Eventual, Inc.)
R Conner Howell is a software engineer at Eventual, Inc. where he works on Daft, a distributed engine for multimodal workloads. He was the original contributor of Daft's SQL implementation, and previously worked on the PartiQL language for AWS Redshift. He joined Eventual, Inc. at the start of 2025 and built out Daft's Catalog integrations for Iceberg, Unity, S3 Tables, and Glue which enabled Daft SQL to query tables from these catalogs. He lives in Seattle and enjoys climbing and cycling in the PNW.
Topic 2: Secure Views for Dynamic Policy Enforcement in Apache Polaris
Abstract
Enforcing fine-grained access control (FGAC) consistently across multiple query engines like Trino, Spark, and StarRocks often requires intrusive, engine-specific changes to evaluate policies. This talk introduces Secure Views for Dynamic Policy Enforcement, an open-source proposal in Apache Polaris—building on prior OSS work by Amazon—that extends Apache Iceberg's view specification to deliver access decisions instead of raw policies. By centralizing policy evaluation in the catalog, engines can uniformly enforce row- and column-level security simply by executing secure views. We'll explore the motivation for enabling multi-engine, cross-cloud governance, the design for avoiding recursive resolution issues, and real-world inspirations like LinkedIn's ViewShift. Attendees will see how this approach makes consistent, low-friction data governance feasible in diverse analytics environments.
Speaker bio
Prashant Singh (Senior Software Engineer @ Snowflake)
Prashant is a committer @ Apache Polaris, contributor @ Apache Iceberg and Apache Spark.
He spent last 5 years integrating Apache Iceberg to various engines of AWS (EMR, Spark, Redshift, Firehose)
Topic 3: What Slows Down Apache Iceberg Queries and How to Fix It
Abstract
This session takes a practical look at what slows Iceberg queries down, from delete file handling to metadata planning overhead, and walks through engineering techniques that address these challenges without sacrificing openness. Topics include optimizing equality delete performance, distributed metadata parsing, vectorized SIMD scans, and more. We'll close with examples from real world deployments and practical techniques you can adopt.
Speaker bio
Sida Shen (PM @ CelerData)
Sida Shen is a contributor to the StarRocks project and a product manager at CelerData. As an engineer with a background in building machine learning and big data infrastructures, he oversees the company's market research while working closely with engineers and developers across the analytics industry to tackle challenges related to data lakehouse analytics.
