📱Build WhatsApp Bot using .NET + OpenAI

  In this article, you will learn how to build a real-world WhatsApp chatbot using ASP.NET Core and OpenAI. This bot can automatically reply to customer messages and can be turned into a SaaS product.

---

🚀 What You Will Build

• WhatsApp webhook API
• AI auto-reply system
• OpenAI integration
• Real-time message processing

---

📌 Prerequisites

• ASP.NET Core Web API
• OpenAI API Key
• WhatsApp Business API (Meta Cloud API or Twilio)

---

1. Setup WhatsApp Cloud API

Go to Meta Developer Portal and create WhatsApp Business app.

https://developers.facebook.com/

Get:

• Access Token
• Phone Number ID
• Webhook Verify Token

---

2. Create ASP.NET Core Web API

dotnet new webapi -n WhatsAppBot

---

3. Configure appsettings.json

"WhatsApp": {
  "AccessToken": "YOUR_TOKEN",
  "PhoneNumberId": "YOUR_PHONE_ID"
},
"OpenAI": {
  "ApiKey": "YOUR_OPENAI_KEY"
}

---

4. Create OpenAI Service

public class OpenAIService
{
    private readonly HttpClient _http;

    public OpenAIService(HttpClient http)
    {
        _http = http;
    }

    public async Task<string> GetReply(string message, string apiKey)
    {
        var request = new
        {
            model = "gpt-4o-mini",
            messages = new[]
            {
                new { role = "user", content = message }
            }
        };

        _http.DefaultRequestHeaders.Authorization =
            new AuthenticationHeaderValue("Bearer", apiKey);

        var response = await _http.PostAsJsonAsync(
            "https://api.openai.com/v1/chat/completions", request);

        dynamic result = await response.Content.ReadFromJsonAsync<dynamic>();

        return result.choices[0].message.content;
    }
}

---

5. Create WhatsApp Webhook Controller

[ApiController]
[Route("api/webhook")]
public class WebhookController : ControllerBase
{
    private readonly IConfiguration _config;
    private readonly OpenAIService _ai;
    private readonly HttpClient _http;

    public WebhookController(IConfiguration config, OpenAIService ai, IHttpClientFactory factory)
    {
        _config = config;
        _ai = ai;
        _http = factory.CreateClient();
    }

    [HttpGet]
    public IActionResult Verify(
        [FromQuery] string hub_mode,
        [FromQuery] string hub_verify_token,
        [FromQuery] string hub_challenge)
    {
        if (hub_verify_token == "my_verify_token")
            return Ok(hub_challenge);

        return Unauthorized();
    }

    [HttpPost]
    public async Task Receive([FromBody] dynamic data)
    {
        try
        {
            var message = data.entry[0].changes[0].value.messages[0].text.body;
            var from = data.entry[0].changes[0].value.messages[0].from;

            var reply = await _ai.GetReply(message, _config["OpenAI:ApiKey"]);

            await SendWhatsAppMessage(from, reply);

            return Ok();
        }
        catch
        {
            return Ok();
        }
    }

    private async Task SendWhatsAppMessage(string to, string message)
    {
        var url = $"https://graph.facebook.com/v18.0/{_config["WhatsApp:PhoneNumberId"]}/messages";

        var payload = new
        {
            messaging_product = "whatsapp",
            to = to,
            text = new { body = message }
        };

        var request = new HttpRequestMessage(HttpMethod.Post, url);
        request.Headers.Authorization =
            new AuthenticationHeaderValue("Bearer", _config["WhatsApp:AccessToken"]);

        request.Content = JsonContent.Create(payload);

        await _http.SendAsync(request);
    }
}

---

6. Register Services

builder.Services.AddHttpClient<OpenAIService>();
builder.Services.AddHttpClient();

---

7. Test Flow

• User sends WhatsApp message
• Webhook receives message
• OpenAI generates reply
• Bot sends reply back

---

💡 Real Use Cases

• Customer support automation
• Order tracking bot
• FAQ chatbot
• Lead generation system

---

💰 Monetization Idea (IMPORTANT)

Turn this into SaaS:

• Charge ₹499/month per business
• Provide dashboard
• Add analytics
• Multi-business support

---

🔐 Best Practices

• Validate webhook requests
• Use logging (Serilog)
• Store chat history
• Add rate limiting
• Use retry policy

---

📌 Conclusion

You now have a real-world WhatsApp AI bot using ASP.NET Core and OpenAI. This can be scaled into a powerful business product.