As AI systems evolve from passive assistants into more autonomous, goal-driven agents, the need for open, reliable, and interoperable infrastructure becomes increasingly important. This panel explores how Apache projects can help empower the next wave of agentic AI by providing the core building blocks for data access, processing, governance, streaming, orchestration, and execution.

From Apache Kafka for real-time event streams, Apache Spark and Apache Flink for large-scale data processing, Apache Iceberg for modern lakehouse storage, to metadata and governance projects such as Apache Gravitino, the Apache ecosystem already provides much of the foundation needed for intelligent agents to interact with enterprise data safely and effectively. The discussion will examine where open-source data infrastructure is already enabling agentic workflows, what technical gaps still remain, and how Apache communities can collaborate to define the next generation of AI-native data systems.

This panel is both a reflection on Apache’s long-standing role in shaping the modern data stack and a forward-looking conversation about how open communities can help build trusted, governed, and extensible foundations for agentic AI.

Speakers:

Junping Du: Found & CEO

Founder and CEO of Datastrato, Ex-Chairperson of LF AI & DATA, ASF Member, Committer and PMC for Apache Hadoop, Co-founder of Apache Ozone, YuniKorn, etc.