Data & AIDeep Dive100min
Integrating AI into Your Java Applications: A Spring AI Deep Dive from Fundamentals to Agentic Patterns
This session explores mastering AI context management with Spring AI—from ChatClient basics, prompt design, and Advisors to RAG integration. It then advances to agentic patterns enabling reasoning, planning, and collaboration among agents, concluding with production-grade observability and security practices for building complete, intelligent Spring-based AI systems.
Christian TzolovBroadcom
talkDetail.whenAndWhere
Wednesday, April 1, 10:50-12:30
Zaal 9
talks.roomOccupancytalks.noOccupancyInfo
The secret to great AI? It's all about context—the art of context window curation. In this hands-on session, we'll explore how Spring AI helps you master this art, taking you from fundamentals to best agentic patterns.
In the first part, we'll cover the fundamentals: ChatClient essentials, prompt engineering, and conversation memory. We'll explore how to extend LLM execution paths using Advisors (intercepting and enhancing AI interactions) and Recursive Advisors (enabling iterative workflows). Enrich your context with RAG, connect to external services using Tools, and use MCP for standardized integration.
The second part focuses on agentic patterns and how to turn an assistant into something that can reason, plan, and act:
- Agent Skills: Modular, LLM-agnostic capabilities loaded on demand
- AskUserQuestion: Gather requirements before acting
- Todo: Prevent "lost in the middle" failures with structured planning
- Subagent Orchestration: Delegate to specialized agents with isolated context windows
- A2A and ACP Protocols: Build interoperable agents that communicate across system boundaries
- Tool Search Tool: Dynamic tool discovery
- LLM-as-a-Judge: Automated response evaluation and quality control
Along the way, we'll address critical production concerns, including observability (metrics, logging, tracing) and security guardrails. You'll leave with a clear understanding of how to integrate generative AI into Spring applications, from your first prompt to a complete agentic architecture
In the first part, we'll cover the fundamentals: ChatClient essentials, prompt engineering, and conversation memory. We'll explore how to extend LLM execution paths using Advisors (intercepting and enhancing AI interactions) and Recursive Advisors (enabling iterative workflows). Enrich your context with RAG, connect to external services using Tools, and use MCP for standardized integration.
The second part focuses on agentic patterns and how to turn an assistant into something that can reason, plan, and act:
- Agent Skills: Modular, LLM-agnostic capabilities loaded on demand
- AskUserQuestion: Gather requirements before acting
- Todo: Prevent "lost in the middle" failures with structured planning
- Subagent Orchestration: Delegate to specialized agents with isolated context windows
- A2A and ACP Protocols: Build interoperable agents that communicate across system boundaries
- Tool Search Tool: Dynamic tool discovery
- LLM-as-a-Judge: Automated response evaluation and quality control
Along the way, we'll address critical production concerns, including observability (metrics, logging, tracing) and security guardrails. You'll leave with a clear understanding of how to integrate generative AI into Spring applications, from your first prompt to a complete agentic architecture
talkDetail.shareFeedback
talkDetail.feedbackNotYetAvailable
talkDetail.feedbackAvailableAfterStart
talkDetail.signInRequired
talkDetail.signInToFeedbackDescription
occupancy.title
occupancy.votingNotYetAvailable
occupancy.votingAvailableBeforeStart
talkDetail.signInRequired
occupancy.signInToVoteDescription
comments.speakerNotEnabledComments