Welcome to the next step in your journey of creating a personal tutor with Spring AI! In the previous lessons, you worked with a ChatService class to send queries to Spring AI's language model, explore model parameters, maintain tutoring session history, and personalize AI behavior with system prompts.

In this lesson, we will generalize and refactor ChatService to a TutoringService to better reflect its educational purpose and to support managing multiple tutoring sessions. This is crucial for applications where you need to handle several educational interactions simultaneously, such as a tutoring platform serving multiple students. By the end of this lesson, you will be able to create and manage multiple tutoring sessions using Spring AI framework, setting the stage for more complex educational interactions.

In a tutoring application, each educational interaction should be treated as a separate session. To achieve this, we use unique identifiers for each tutoring session. This ensures that queries and explanations are correctly associated with their respective sessions.

We'll generate a unique session identifier using UUID.randomUUID().toString(). Each session's conversation history is managed by associating the session ID with its own context, allowing the AI to maintain continuity for each student.

Below is an example of how to create this method:

We'll use this session ID to keep track of the conversation history and context for each student.

By managing sessions in this way, you ensure that each educational interaction remains distinct and coherent.

Once a tutoring session is created, you can send queries and receive explanations while keeping the conversation context intact. To maintain this context across messages, each question must be associated with its corresponding session ID.

In Spring AI, this can be done by passing the session ID to the chat advisor, which ensures the right memory history is retrieved and updated for that specific session. Advisors are a chain of interceptors that can process and modify chat requests and responses, allowing you to add functionalities such as managing conversation history, enforcing usage limits, logging interactions, and more. When you use the .advisors() method, you specify which advisors should be applied to a particular chat prompt, enabling features like session management and context preservation for each conversation.

Below is the updated askQuestion method in TutoringService, which takes a sessionId and uses it to manage memory isolation:

Here’s what’s happening in the advisors configuration:

• ChatMemory.CONVERSATION_ID is a special key that tells Spring AI which session’s message history to use.
• The sessionId you provide acts as the unique identifier for that tutoring session.
• Spring AI automatically retrieves past messages for that session and appends the latest user and assistant messages after each call.

By using a unique sessionId per session, each conversation stays separate and maintains its own memory, ensuring accurate and personalized interactions.

The code below demonstrates how to create and interact with separate tutoring sessions using the TutoringService. Each session maintains its own memory, ensuring that conversations remain isolated and context-specific. This prevents any overlap or interference between different users or topics.

Inside the below example, we simulate two independent sessions by generating unique session IDs and asking different sets of questions in each one:

Sample Output:

This setup ensures that each session's memory is kept separate, allowing multiple conversations to run in parallel without sharing context. It’s a reliable way to provide accurate, personalized answers to different users or topics independently.

In this lesson, you learned how to manage multiple tutoring sessions using the Spring AI framework. We refactored ChatService into a more flexible TutoringService, introduced session-specific memory using unique identifiers, and demonstrated how to handle multiple conversations independently. These capabilities are essential when building intelligent educational tools that need to keep track of multiple learners at once.

As you move on to the practice exercises, try creating and managing separate tutoring sessions on your own. This hands-on experience will help solidify your understanding and prepare you to build more advanced and responsive educational assistants. Keep going—you’re on your way to building a fully personalized tutor with Spring AI!