Basic Chatbot

Build your first AI chatbot with BroxiAI in under 10 minutes

Learn how to create a simple but powerful AI chatbot using BroxiAI's visual workflow builder.

What You'll Build

A conversational AI chatbot that can:

• Respond to user messages naturally

• Maintain conversation context

• Handle various types of questions

• Provide helpful and engaging responses

Prerequisites

• BroxiAI account (sign up at use.broxi.ai)

• OpenAI API key (or another supported AI provider)

• 10 minutes of your time

Step 1: Set Up Your Workspace

1. Log into BroxiAI

   • Navigate to use.broxi.ai

   • Sign in with your credentials

2. Create a New Flow

   • Click "New Flow" from your dashboard

   • Name your flow "My First Chatbot"

   • Select "Blank Flow" to start from scratch

3. Configure API Keys

   • Go to Settings → API Keys

   • Add your OpenAI API key

   • Save the configuration

Step 2: Add Core Components

Chat Input Component

1. Drag Chat Input from the Components sidebar

2. Configure the component:

   {
     "name": "User Message",
     "placeholder": "Type your message here...",
     "session_id": "user_session"
   }

AI Model Component

1. Drag OpenAI Model from the AI Models section

2. Configure the model:

   {
     "model": "gpt-3.5-turbo",
     "temperature": 0.7,
     "max_tokens": 500,
     "system_message": "You are a helpful and friendly AI assistant. Provide clear, concise, and helpful responses to user questions."
   }

Chat Output Component

1. Drag Chat Output from the Components sidebar

2. Configure for display:

   {
     "name": "AI Response",
     "show_in_playground": true
   }

Step 3: Connect the Components

1. Connect Chat Input to OpenAI Model

   • Drag from the output handle (⚪) of Chat Input

   • Connect to the "input" port of the OpenAI Model

2. Connect OpenAI Model to Chat Output

   • Drag from the output handle of OpenAI Model

   • Connect to the input port of Chat Output

Your workflow should look like this:

Step 4: Test Your Chatbot

1. Open the Playground

   • Click the "Playground" button in the top-right corner

   • The chat interface will appear

2. Start Chatting

   • Type a message like "Hello, how are you?"

   • Press Enter or click Send

   • Watch your AI respond!

3. Try Different Prompts

   Examples to try:
   - "What's the weather like?" 
   - "Tell me a joke"
   - "Explain quantum computing"
   - "Help me write an email"

Step 5: Enhance Your Chatbot

Add Memory for Context

1. Add Conversation Memory

   • Drag "Conversation Buffer Memory" component

   • Place it between Chat Input and OpenAI Model

2. Configure Memory:

   {
     "memory_key": "chat_history",
     "max_token_limit": 2000,
     "return_messages": true
   }

3. Update Connections

   • Connect Chat Input → Memory → OpenAI Model → Chat Output

Customize the Personality

Update the system message in your OpenAI Model:

You are Alex, a knowledgeable and friendly AI assistant who works for BroxiAI. 

Your personality:
- Enthusiastic about AI and technology
- Always helpful and patient
- Uses a conversational tone
- Provides practical advice
- Asks follow-up questions when helpful

Remember to:
- Keep responses concise but informative
- Use examples when explaining concepts
- Be encouraging and supportive
- Maintain a professional yet friendly demeanor

Add Response Validation

1. Add Text Processor component after OpenAI Model

2. Configure validation:

   {
     "max_length": 1000,
     "remove_profanity": true,
     "format_markdown": true
   }

Step 6: Deploy Your Chatbot

Publish Your Workflow

1. Test Thoroughly

   • Try various input types

   • Verify responses are appropriate

   • Check conversation flow

2. Publish the Flow

   • Click "Publish" in the API Access panel

   • Copy your API endpoint

   • Note your flow ID

Integration Options

Web Integration

<script>
async function sendMessage(message) {
  const response = await fetch('https://api.broxi.ai/v1/flows/YOUR_FLOW_ID/run', {
    method: 'POST',
    headers: {
      'Authorization': 'Bearer YOUR_API_TOKEN',
      'Content-Type': 'application/json'
    },
    body: JSON.stringify({
      input: message,
      session_id: 'user_123'
    })
  });
  
  const result = await response.json();
  return result.output;
}
</script>

Python Integration

import requests

def chat_with_bot(message, session_id="default"):
    response = requests.post(
        "https://api.broxi.ai/v1/flows/YOUR_FLOW_ID/run",
        headers={
            "Authorization": "Bearer YOUR_API_TOKEN",
            "Content-Type": "application/json"
        },
        json={
            "input": message,
            "session_id": session_id
        }
    )
    
    return response.json()["output"]

# Usage
response = chat_with_bot("Hello, how can you help me?")
print(response)

Advanced Features

Add Knowledge Base

1. Upload Documents

   • Use File Manager to upload relevant documents

   • Add Document Loader component

   • Connect to Text Splitter

2. Create Vector Database

   • Add Vector Database component

   • Configure embeddings

   • Store document chunks

3. Implement RAG

   • Add Retrieval component

   • Connect to your vector database

   • Enhance model with retrieved context

Multi-Language Support

1. Add Language Detection

   • Insert Language Detector component

   • Identify user's language automatically

2. Add Translation

   • Use Translation component

   • Translate responses to user's language

Analytics and Monitoring

1. Track Conversations

   • Log user interactions

   • Monitor response quality

   • Analyze usage patterns

2. Set Up Alerts

   • Monitor error rates

   • Track response times

   • Set up usage notifications

Troubleshooting

Common Issues

Bot Not Responding

• Check API key configuration

• Verify component connections

• Review error logs in the Playground

Poor Response Quality

• Adjust temperature settings

• Improve system message

• Add more context or examples

Memory Issues

• Check memory configuration

• Adjust token limits

• Clear conversation history if needed

Next Steps

Congratulations! You've built your first BroxiAI chatbot. Here's what to explore next:

1. Advanced Components

   • Agents with Tools

   • Vector Databases

   • Custom Tools

2. Integration Examples

   • Slack Bot

   • Website Chat Widget

   • WhatsApp Business

3. Best Practices

   • Performance Optimization

   • Security Guidelines

   • Cost Management

Resources

• Template: Download the Basic Chatbot Template

• Video Tutorial: Building Your First Chatbot

• Community: Share your chatbot in our Discord

• Support: Get help in our Community Forums

---

You've successfully created your first AI chatbot! This is just the beginning of what's possible with BroxiAI. Experiment with different components and configurations to create more sophisticated applications.