Handling AI Interactions with the Chat Service Layer in JavaScript

Handling AI Interactions with the Chat Service Layer

In the previous lesson, we explored the ChatManager class, which plays a crucial role in managing chat data within our application. Now, we will take the next step in our journey by building the Chat Service Layer . This layer is essential for integrating the language model with chat sessions, allowing us to process user messages and generate AI responses. By the end of this lesson, you will understand how to set up the ChatService class, create chat sessions, and process messages using OpenAI's API. The service layer acts as a bridge between the model layer, where data is managed, and the AI model, which generates responses. It is responsible for orchestrating the flow of data and ensuring that user interactions are handled smoothly. Let's dive into the details of setting up this important component.

Setting Up the ChatService Class

The ChatService class is the heart of our service layer. It is responsible for managing chat sessions and interacting with the OpenAI client to generate AI responses. To begin, we need to set up the class and its components. First, we import the necessary modules, including the ChatManager from our previous lesson and the OpenAI library for API requests. We also use the uuid package to generate unique chat IDs. Here's how the class is initialized: JavaScriptimport { v4 as uuidv4 } from 'uuid'; import { OpenAI } from 'openai'; import ChatManager from './models/chat.js'; class ChatService { constructor() { this.chatManager = new ChatManager(); this.systemPrompt = this.loadSystemPrompt('data/system_prompt.txt'); this.openai = new OpenAI({ apiKey: process.env.OPENAI_API_KEY, }); } } In this setup, we instantiate ChatManager to manage chat data, load the systemPrompt using the loadSystemPrompt method, and initialize the OpenAI client with an API key from the environment variables.

Loading the System Prompt

The system prompt is a crucial component that guides the AI's responses. It provides context and instructions for the AI, ensuring that it behaves in a manner consistent with our application's goals. In this section, we'll implement the loadSystemPrompt method to load the prompt from a file. JavaScript import fs from 'fs'; loadSystemPrompt(filePath) {try {return fs.readFileSync(filePath, 'utf8');} catch (error) {console.error(`Error loading system prompt: ${error}`); return "You are a helpful assistant.";}} This method attempts to read the system prompt from a specified file path. If successful, it returns the prompt as a string. In case of an error, it prints an error message and returns a default prompt. This ensures that the application can continue functioning even if the file is missing or corrupted.

Join the 1M+ learners on CodeSignal

Be a part of our community of 1M+ users who develop and demonstrate their skills on CodeSignal

Processing User Messages

The processMessage method is where the magic happens. It processes user messages, interacts with the OpenAI client to generate AI responses, and updates the chat history. Below, we outline the steps involved in this process, followed by the corresponding code implementation:

Retrieve the chat using getChat, and raise an error if the chat is not found.
Add the user's message to the chat history.
Send the conversation, including the system prompt and all messages, to the OpenAI client to generate a response.
Add the AI's response to the chat history and return it to the user.
Handle any errors with the AI client gracefully.

JavaScript

async processMessage(userId, chatId, message) {
    // Step 1: Retrieve the chat
    const chat = this.chatManager.getChat(userId, chatId);
    if (!chat) {
        throw new Error("Chat not found");
    }

    // Step 2: Add user message to chat history
    this.chatManager.addMessage(userId, chatId, "user", message);

    try {
        // Step 3: Get AI response
        const conversation = this.chatManager.getConversation(userId, chatId);

        const response = await this.openai.chat.completions.create({
            model: "gpt-6-luna",
            messages: conversation,
            max_completion_tokens: 500
        });

        const aiMessage = response.choices[0].message.content;

        // Step 4: Add AI response to chat history
        this.chatManager.addMessage(userId, chatId, "assistant", aiMessage);

        return aiMessage;

    } catch (error) {
        // Step 5: Handle errors
        throw new Error(`Error getting AI response: ${error.message}`);
    }
}

async processMessage(userId, chatId, message) {
    // Step 1: Retrieve the chat
    const chat = this.chatManager.getChat(userId, chatId);
    if (!chat) {
        throw new Error("Chat not found");
    }

    // Step 2: Add user message to chat history
    this.chatManager.addMessage(userId, chatId, "user", message);

    try {
        // Step 3: Get AI response
        const conversation = this.chatManager.getConversation(userId, chatId);

        const response = await this.openai.chat.completions.create({
            model: "gpt-6-luna",
            messages: conversation,
            max_completion_tokens: 500
        });

        const aiMessage = response.choices[0].message.content;

        // Step 4: Add AI response to chat history
        this.chatManager.addMessage(userId, chatId, "assistant", aiMessage);

        return aiMessage;

    } catch (error) {
        // Step 5: Handle errors
        throw new Error(`Error getting AI response: ${error.message}`);
    }
}

In the context of a customer service agent, we configure our model with a limit on the length of its answers. The max_completion_tokens parameter is set to 500, allowing the model to provide detailed and informative answers without overwhelming the user, thus maintaining a smooth and effective customer service experience. Note that we don't set temperature here: newer models like gpt-6-luna manage their own sampling and only accept its default value.

Example: Simulating a Chat Session

Let's see the ChatService in action by simulating a chat session. We'll use a main.js file to create a chat session and process a user message.

JavaScript

import ChatService from './services/chat_service.js';

// Initialize the chat service
const chatService = new ChatService();

// Simulate a user ID
const userId = "user123";

// Create a new chat session
const chatId = chatService.createChat(userId);
console.log(`Chat session created with ID: ${chatId}`);

// Simulate sending a message
const userMessage = "Hello, how are you?";

(async () => {
    try {
        const aiResponse = await chatService.processMessage(userId, chatId, userMessage);
        console.log(`AI Response: ${aiResponse}`);
    } catch (error) {
        console.error(`Error: ${error.message}`);
    }
})();

import ChatService from './services/chat_service.js';

// Initialize the chat service
const chatService = new ChatService();

// Simulate a user ID
const userId = "user123";

// Create a new chat session
const chatId = chatService.createChat(userId);
console.log(`Chat session created with ID: ${chatId}`);

// Simulate sending a message
const userMessage = "Hello, how are you?";

(async () => {
    try {
        const aiResponse = await chatService.processMessage(userId, chatId, userMessage);
        console.log(`AI Response: ${aiResponse}`);
    } catch (error) {
        console.error(`Error: ${error.message}`);
    }
})();

In this example, we initialize the ChatService, simulate a user ID, and create a new chat session, printing the chat ID. We then simulate sending a message and print the AI's response, demonstrating the flow from user input to AI response and showcasing the functionality of the ChatService.

text

Chat session created with ID: 01a17870-8a4f-4b6f-a3ce-f04e1136d597
AI Response: Hello! I'm here to help with any questions or concerns you might have regarding our IT services. How can I assist you today?

Chat session created with ID: 01a17870-8a4f-4b6f-a3ce-f04e1136d597
AI Response: Hello! I'm here to help with any questions or concerns you might have regarding our IT services. How can I assist you today?

This output illustrates a successful interaction where a new chat session is created, and the AI responds to the user's greeting with a helpful message. The AI's response is tailored to assist with IT services, showcasing the system's ability to provide relevant and context-aware assistance.