What is an Agentic Application?

2026-05-06 · Source: LLM on Medium · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering · Depth: Intermediate, quick

Summary

This article details the construction of a simple agentic application using the Spring AI framework, specifically focusing on its integration with AWS Bedrock for Large Language Model (LLM) capabilities. The application is designed to interact with an MCP server, also built with Spring AI, which exposes tools via an HTTP streaming protocol in a stateless mode. For conversational memory, it employs Spring AI's built-in InMemory repository and MessageWindowChatMemory. The practical scenario demonstrated involves the application providing weather information and latitude/longitude data for specified cities, showcasing Spring AI's capabilities in orchestrating LLM interactions, tool usage, and memory management within an agentic architecture.

Key takeaway

For AI Engineers developing agentic applications, leveraging Spring AI offers a structured approach to integrate LLMs like AWS Bedrock, manage conversational memory, and expose tools. You should consider using its built-in memory repositories and stateless MCP server integration to streamline development and ensure robust interaction capabilities for your applications.

Key insights

Spring AI facilitates building agentic applications by integrating LLMs, tools, and memory components.

Principles

• Stateless tool exposure enhances scalability.
• In-memory chat history supports conversational context.

Method

Build agentic applications by connecting Spring AI to LLMs (e.g., AWS Bedrock), exposing tools via an MCP server, and managing conversation history with MessageWindowChatMemory.

In practice

• Integrate AWS Bedrock for LLM access.
• Use Spring AI for tool exposure via HTTP.
• Implement InMemoryChatMemory for context.

Topics

• Agentic Applications
• Spring AI
• AWS Bedrock
• MCP Server
• Chat Memory

Best for: AI Engineer, Software Engineer, AI Student

Related on AIssential

• Introducing stateful MCP client capabilities on Amazon Bedrock AgentCore Runtime — Stateful MCP on Amazon Bedrock AgentCore Runtime enables interactive, multi-turn AI agent workflows.
• Beyond the Chatbot: How AI Learned to Think, Plan, and Act for You — Agentic AI combines LLMs with planning and external tools to execute complex, multi-step tasks autonomously.
• Gemini Interactions API Quick Start — The Gemini Interactions API unifies agentic application development with server-side state, tool orchestration, and multimodal capabilities.
• Building a Tool-Augmented RAG Agent with Session Memory — Tool-augmented RAG agents with session memory enable multi-turn, context-aware conversations by dynamically calling a knowledge base.
• From Tool Calling to MCP: Building a Natural Language Search with Spring AI and MCP Server — MCP standardizes AI model communication with external tools, enabling reusability and independent deployment.
• Agentic AI in Action — Part — 22 — Memory in Agentic AI on Snowflake — Memory transforms AI from a reactive tool to an adaptive, continuously learning participant by enabling experience accumulation.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by LLM on Medium.