Confluent (CFLT) Current 2025 Day 2 Keynote summary

Opening keynote and historical context

• Emphasized the importance of information velocity, drawing parallels from the defeat of the Spanish Armada to modern real-time systems.

• Traced the evolution from monolithic architectures to microservices, highlighting agility and business alignment.

• Discussed the emergence of Kafka as a central nervous system for data, and the challenges of early data lakes.

Data architecture evolution and streaming

• Outlined the shift from batch processing with Hadoop and Spark to modern table formats like Iceberg, Hudi, and Delta Lake.

• Stressed the move from data swamps to reliable, governed data lakes with ACID guarantees and time travel.

• Noted the default adoption of streaming and the need for real-time computation as events occur.

AI integration and system readiness

• Highlighted the complexity AI brings, requiring new layers for auditability, logs, guardrails, and real-time context.

• Asserted that AI must be built on event-driven, real-time principles to be trustworthy and effective.

• Encouraged active audience participation in capturing real-time events, feeding into an AI-powered dashboard.