Spring AI 是一个面向人工智能工程的应用框架。解决了 AI 集成的基本挑战将企业数据和API与AI 模型连接起来。spring ai: https://docs.spring.io/spring-ai/reference/spring ai的特性提示词工厂可以说是大模型应用中最简单也是最核心的一个技术。他是我们更大模型交互的媒介提示词给的好大模型才能按你想要的方式响应。对话拦截advisors对话记忆Autowired ChatMemoryRepository chatMemoryRepository;注通过一个bean组件就可以让大模型拥有对话记忆功能toolsclass DateTimeTools { Tool(description Get the current date and time in the users timezone) String getCurrentDateTime() { return LocalDateTime.now().atZone(LocaleContextHolder.getTimeZone().toZoneId()).toString(); } }RAG技术下的 ETL让大模型可以跟企业业务数据进行互联包括读取文件、分隔文件、向量化 向量数据库支持 目前支持20种向量数据库的集成MCP让tools外部化形成公共工具让外部开箱即用。agent应用Evaluator Optimizer – The model analyzes its own responses and refines them through a structured process of self-evaluation.Routing – This pattern enables intelligent routing of inputs to specialized handlers based on classification of the user request and context.Orchestrator Workers – This pattern is a flexible approach for handling complex tasks that require dynamic task decomposition and specialized processingChaining – The pattern decomposes complex tasks into a sequence of steps, where each LLM call processes the output of the previous one.Parallelization – The pattern is useful for scenarios requiring parallel execution of LLM calls with automated output aggregation. SpringAI 1.0 实战 ：认识spring ai
Spring AI 是一个面向人工智能工程的应用框架。解决了 AI 集成的基本挑战将企业数据和API与AI 模型连接起来。spring ai: https://docs.spring.io/spring-ai/reference/spring ai的特性提示词工厂可以说是大模型应用中最简单也是最核心的一个技术。他是我们更大模型交互的媒介提示词给的好大模型才能按你想要的方式响应。对话拦截advisors对话记忆Autowired ChatMemoryRepository chatMemoryRepository;注通过一个bean组件就可以让大模型拥有对话记忆功能toolsclass DateTimeTools { Tool(description Get the current date and time in the users timezone) String getCurrentDateTime() { return LocalDateTime.now().atZone(LocaleContextHolder.getTimeZone().toZoneId()).toString(); } }RAG技术下的 ETL让大模型可以跟企业业务数据进行互联包括读取文件、分隔文件、向量化 向量数据库支持 目前支持20种向量数据库的集成MCP让tools外部化形成公共工具让外部开箱即用。agent应用Evaluator Optimizer – The model analyzes its own responses and refines them through a structured process of self-evaluation.Routing – This pattern enables intelligent routing of inputs to specialized handlers based on classification of the user request and context.Orchestrator Workers – This pattern is a flexible approach for handling complex tasks that require dynamic task decomposition and specialized processingChaining – The pattern decomposes complex tasks into a sequence of steps, where each LLM call processes the output of the previous one.Parallelization – The pattern is useful for scenarios requiring parallel execution of LLM calls with automated output aggregation.