Built a Multimodal AI Executive Assistant on WhatsApp using n8n
I wanted to build an AI assistant that can actually interact with people naturally — not just through text.
So I built a WhatsApp-based AI Agent that can handle:
💬 Text messages
🎙️ Voice messages
🖼️ Images
🧠 Conversation memory
🔎 Web search
📅 Google Calendar
📧 Gmail
📊 Google Sheets
The interesting part?
The user can send a voice message → the system transcribes it → the AI understands it → takes the required action → and can respond back with text or voice.
For images, the workflow analyzes the image and combines its content with the user's message before sending it to the AI Agent.
I also connected the agent with external tools so it can do more than just answer questions — it can search, manage calendar events, send emails, and retrieve information from a database.
Built with:
⚙️ n8n
🤖 Google Gemini
📱 WhatsApp Cloud API
🧠 AI Agent + Memory
🔊 Kokoro TTS
📅 Google Calendar
📧 Gmail
📊 Google Sheets
This project pushed me further from simply building workflows to designing AI agents that can understand different types of input and interact with real-world tools.
The goal: AI that doesn't just talk — it gets things done.
#n8n #AIAutomation #AIAgents #WorkflowAutomation #WhatsAppAutomation #GenerativeAI #GoogleGemini #Automation