Development of AI chat assistants with Java/Spring AI

This course provides a comprehensive insight into the implementation of AI chat assistants with the Retrieval Augmented Generation pattern using the latest Java Spring library. With a mix of theoretical explanations and practical exercises, you will learn to develop and customize advanced AI chat assistants.

You will learn how the retrieval-augmented generation pattern for AI text generation works. Basic concepts such as models, embeddings and vector-based databases are introduced. In addition, the use of prompting techniques and the interaction between retrieval and generation are discussed. The pattern uses a two-step approach in which relevant text fragments are first retrieved from a vector-based database and then an answer is formulated using generation models. Emphasis is placed on both the efficiency of the query and the quality of the generation.

The user interface will be developed using Angular as an example to learn best practices for handling streamed AI responses.

The course will be delivered through a combination of lectures, practical exercises, discussions and demos. In addition, you will also gain access to online resources, including the official Spring AI documentation, to deepen your understanding and complete additional exercises.

• Spring AI is a new addition to the Spring framework that allows us to work with AI chat models in enterprise applications.
• Introduction to the theoretical background of AI models: basic concepts and principles behind the development of AI chat assistants are presented. This includes an overview of various AI techniques such as machine learning, natural language processing and neural networks.
• Fundamentals of Spring AI: Introduction to the abstract concepts that serve as the foundation for developing AI applications.
• Core abstractions: Explanation of Spring AI's core abstractions that enable different implementations and allow for easy replacement of components with minimal code changes. The ChatClient /StreamingChatClient interfaces and their implementations for OpenAI, Azure OpenAI, Ollama, VertexAI, Huggingface, Bedrock/Llama2, Bedrock/Anthropic, Bedrock/Titan, Bedrock/Cohere as well as abstractions such as EmbeddingClient and ImageClient and their model implementations are covered.
• Higher-level functionalities: Presentation of Spring AI's higher level functionalities to handle common use cases such as "Questions and answers about your documentation" or "Chatting with your documentation".
• Integration with Spring ecosystem projects: Explanation of how Spring AI can be integrated with other projects in the Spring ecosystem, such as Spring Integration, Spring Batch, Spring Data, Spring Cloud GCP, Spring Cloud, etc.
• Setup Simplification: Using Spring Boot Starters to simplify the setup of key dependencies and classes, and showing sample applications to explore the features of the project.
• Using Spring CLI: Introduction to the new Spring CLI project that allows you to get started quickly by using the "spring boot new ai" command for new projects or "spring boot add ai" to add AI functionality to an existing application.

• Software developers and architects

You will learn how to integrate AI components into custom software with Spring AI in a practical way. The focus is on the retrieval-augmented generation pattern, the implementation of generative AI functions, and the reliable integration of LLMs into Java applications.