Human-AI Collaboration: Making Prudent Use of AI in Development - Venkat Subramaniam

This talk addresses the challenges faced by Java developers as AI transitions from experimentation to production, specifically in the context of building intelligent, context-aware systems. The speaker explores the evolution of AI-native Java applications, emphasizing agent-based architecture and the transformative effects of the Model Context Protocol (MCP) and Agent-to-Agent (A2A) communication patterns. Attendees will understand how to transition from single-turn interactions to multi-agent coordination, persistent memory, and tool-augmented reasoning. Additionally, the session delves into architectural trade-offs related to performance, flexibility, and governance, while highlighting the importance of security and observability in developing modern intelligent services. As enterprises adopt AI-native platforms, the talk positions Java as a mature and robust choice for this innovative software landscape.