Unlock the Power of Open Data Formats for
AI-Ready Analytics with Iceberg

View the webinar presentation files — covering why open data formats matter, the shift from proprietary lock-in to open lakehouses, and how Apache Iceberg on AWS enables AI-ready analytics at scale.

Traditional data lakes lack ACID transactions, struggle with small files, and make updates/deletes complex. The result is data teams spending more time fixing infrastructure than enabling AI. Apache Iceberg addresses these challenges with a metadata-driven table format that enables transactions, time travel, schema/partition evolution, and compaction without data rewrites. This session covers Iceberg's core concepts, its 4-layer metadata architecture, and how AWS managed services (S3 Tables, Glue Iceberg Optimizer) automate table maintenance enabling you to move from a traditional data lake to a fully transactional lakehouse.

Learn how to build a governed, production-ready data lakehouse using Amazon S3 Tables — fully managed Apache Iceberg storage with zero operational overhead, 3×faster queries, and automatic cost optimization. This session covers ingestion patterns (Zero-ETL, streaming, batch), data governance (catalog, lineage, quality, fine-grained access), and multi-engine consumption including generative AI.

A live demo showcasing how RetailPulse builds an AI-ready data foundation with Apache Iceberg — combining real-time lakehouse architecture with agentic AI to deliver intelligent, autonomous analytics.

Data loses 50% of its value within 8 hours, yet batch pipelines still introduce hours-to-days of latency. This session covers how to build real-time streaming architectures using Apache Iceberg on AWS.

The Bronze layer is where your AI-ready lakehouse begins. By landing every source — batch and streaming — in open Apache Iceberg format, you capture the raw, trustworthy signal that everything downstream (analytics and, ultimately, agents) depends on. No AI outcome is better than the data beneath it.

AI and agents can't reason over fragmented data. The Silver layer resolves that dysfunction by merging batch profiles with streaming transactions into one consistent, ACID-guaranteed source of truth.

The Gold layer is the business-ready foundation an AI agent reasons over. Pre-aggregated, well-defined Customer 360 metrics give the semantic layer clean, unambiguous concepts to ground on — so when a stakeholder asks a natural-language question, the agent returns a governed, accurate answer instead of a guess.

Connect Iceberg tables to Amazon QuickSight to build interactive dashboards surfacing fraud anomalies, spending trends, and customer segments without data copies.

This session covers Amazon Redshift's Apache Iceberg integration, highlighting how Redshift enables high-performance SQL queries on lakehouse data. The demonstration shows practical implementation, including querying Iceberg tables and joining with local data warehouse tables as well as Redshift's materialized view support.

This session covers practical strategies for migrating to Apache Iceberg on AWS from data lakes, other open table formats (Hudi, Delta), self-managed Iceberg environments, and data warehouses. Attendees will learn when to use in-place migration versus full data migration, how to plan cutover approaches (direct vs. phased), and how to maintain Iceberg tables at scale using AWS-managed solutions like Amazon S3 Tables and AWS Glue Iceberg Optimizer.