n8n WhatsApp Auto-Response Integration
Budget: ₹12,500 – ₹37,500 INR
I want to build an n8n workflow that takes every incoming WhatsApp message, routes it through an LLM layer (OpenAI GPT-5 or Claude orchestrated with LangGraph / CrewAI and following MCP standards), enriches the prompt with Retrieval-Augmented Generation against our internal data, then sends an intelligent automated reply back through the WhatsApp Business API.
The immediate milestone is the WhatsApp piece, but the architecture must stay flexible so I can later plug in Google Workspace and CRM data sources without rewriting the flow. I already run Python + FastAPI services, so any custom code you create should expose endpoints that slot straight into that stack.
Key deliverables
• Ready-to-run n8n workflow: trigger on WhatsApp, call LLM + RAG service, return response
• Lightweight FastAPI service that handles model calls and retrieval logic
• Environment and credential setup instructions so I can replicate everything on my own server
• Short Loom or text walkthrough showing the flow live and explaining how to extend it for Google Workspace or CRM hooks
The WhatsApp API access is in place; bring your n8n, Node.js, Python and LLM integration experience so we can get this running quickly and cleanly.
The immediate milestone is the WhatsApp piece, but the architecture must stay flexible so I can later plug in Google Workspace and CRM data sources without rewriting the flow. I already run Python + FastAPI services, so any custom code you create should expose endpoints that slot straight into that stack.
Key deliverables
• Ready-to-run n8n workflow: trigger on WhatsApp, call LLM + RAG service, return response
• Lightweight FastAPI service that handles model calls and retrieval logic
• Environment and credential setup instructions so I can replicate everything on my own server
• Short Loom or text walkthrough showing the flow live and explaining how to extend it for Google Workspace or CRM hooks
The WhatsApp API access is in place; bring your n8n, Node.js, Python and LLM integration experience so we can get this running quickly and cleanly.